Return-Path: <maiser@efs.mq.edu.au>
Received: from corot.bc.edu (corot.bc.edu [136.167.2.209])
	by monet.bc.edu (8.8.7/8.8.7) with ESMTP id MAA266372
	for <baum@monet.bc.edu>; Fri, 2 Apr 1999 12:22:51 -0500
From: maiser@efs.mq.edu.au
Received: (from root@localhost)
	by corot.bc.edu (8.8.7/8.8.7) with X.500 id MAA91472
	for baum@mail1.bc.edu; Fri, 2 Apr 1999 12:22:49 -0500
Received: from sunb.ocs.mq.edu.au (sunb.ocs.mq.edu.au [137.111.1.11])
	by corot.bc.edu (8.8.7/8.8.7) with ESMTP id MAA114942
	for <baum@bc.edu>; Fri, 2 Apr 1999 12:22:35 -0500
Received: from efs1.efs.mq.edu.au (efs1.efs.mq.edu.au [137.111.64.8])
	by sunb.ocs.mq.edu.au (8.9.2/8.9.2) with ESMTP id DAA26625
	for <baum@bc.edu>; Sat, 3 Apr 1999 03:22:27 +1000 (EST)
Received: from EFS1/SpoolDir by efs1.efs.mq.edu.au (Mercury 1.40);
    3 Apr 99 03:24:15 GMT+1000
Received: from SpoolDir by EFS1 (Mercury 1.40); 3 Apr 99 03:24:14 GMT+1000
To: baum@bc.edu
Date: Sat, 3 Apr 99 3:24:12 GMT+1000
Subject: Re: 
Message-ID: <82554076832@efs1.efs.mq.edu.au>

From: "Iacoviello,M" <M.Iacoviello@lse.ac.uk>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: RE: Norman Morin's VAR.SRC
Date: Sun, 31 Jan 1999 17:49:43 -0000
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2232.9) (via Mercury MTS (Bindery) v1.40)
Content-Type: text/plain

Dear Norman,
I managed to perform the Blanchard-Quah decomposition without big problems,
although the only way to identify the shocks is to assume a recursive
structure in the long-run coefficients matrix. (this is what I read in the
description of the program). For instance, I had a dataset of quarterly
observations going from 1982 to 1997, and this is what I did. You should
have no problems, even though I had to try many many times (probably because
I have only Rats4.2 and - even if I am a new user - I dont find it very
user-friendly).
These were my instructions



calendar 1982 1 4
all 80 97:2 ; * these were my observations
ope c:\winrats\uk.rat * this was my file
data(for=rats, org=obs)


source(noecho) c:\winrats\var.src * this was the routine

@VAR 1982:3 1997:2 B0 * this to run the var
# y m p b
# constant;


****************************************
*   Vector  Autoregressive  Analysis   *
****************************************

and everything should work fine. At the end, I also got confirmation that
everything was fine because the Accumulated Impulse Responses were showing
that the long-run effects for some shocks were zero in a triangular fashion
- see below.

Accumulated Impulse Responses
    Normalized so each structural form shock = one standard deviation
    (i,j)th element is the accumulated effect
         on variable i of a shock to variable j
  Out to 18 steps
     Y           M           P           B
     0.03528    -0.00292    -0.00170    -0.00014
     0.05305     0.07829    -0.00024    -0.00042
     0.00256     0.03276     0.01119     0.00191
     0.17539     0.16760     0.05058     0.05506

  Out to oo steps
     Y           M           P           B
     0.02444    -0.00000    -0.00000    -0.00000 * the order was Y M P H.
     0.07014     0.09744     0.00000    -0.00000
     0.01182     0.03350     0.00835     0.00000
     0.14647     0.17098     0.04416     0.04543

Yours sincerely,

Matteo

****************************************************

Matteo Iacoviello
Research students - Dept of Economics
London School of Economics
Houghton Street
London WC2A 2AE


> ----------
> From: 	Van & Bethanie Newby[SMTP:newby@micron.net]
> Reply To: 	RATS Discussion List
> Sent: 	Saturday, 30 January 1999 10:20
> To: 	RATS Discussion List
> Subject: 	Norman Morin's VAR.SRC
> 
> Hello,
> 
> I want to use RATS to perform a Blanchard and Quah decomposition.
> Fortunately, Norman Morin has recently included this in VAR.SCR.  However,
> I am having many problems using it.  While I can set the deterministic
> component and lag length, and estimate the VAR, I get errors everytime I
> attempt to graph the ACF/PACF.  Furthermore, I get errors when trying to
> run the Blanchard and Quah decomp part of the program (within the impulse
> response analysis).  I'm not sure how to include the necessary
> information.
> Thanks for the help.
> 
> Van
> 

---------- End of message ----------

From: Ricardo Chaves Lima <rcl@netpe.com.br>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: generation data
Date: Sun, 31 Jan 1999 21:48:25 -0300
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
X-Mailer: Mozilla 4.01 [en] (Win95; I) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Dear fellows,

Does anyone kwno how to generate data (simulate) on RATS.
I would like to ganerate 100 observation of the following model:

               y(t)  =  0.8*y(t-1) + e

being "e" a random error N ~ (0, 2).

thanks

Ricardo Chaves Lima
rcl@netpe.com.br


---------- End of message ----------

From: Karin Elsner <elsner@iamo.uni-halle.de>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: RATS for estimation of AIDS using cross-section data
Date: Mon, 01 Feb 1999 09:39:45 +0100
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Organization: Institut für Agrarentwicklung
X-Mailer: Mozilla 3.01 [de] (Win95; I) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Dear colleagues,
I am using RATS for doing cross-sectional analysis because this software
is available in the institute I am working at. More specifically, I
estimate an Almost Ideal Demand System including socio-demographic
variables.
I face a lot of problems in transferring the algebraic model and the
estimation procedures necessary for cross-section analysis (e.g.
modified Heckamn procedure to account for zero expenditures, and so on)
into the RATS programme syntax. In order to avoid inventing the wheel a
hundredth time, I would like to ask whether somebody has estimated an
AIDS with RATS using cross-sectional or also time-series data and
whether he or she could imagine giving me the respective programme file.
I know that this file would in no case be self-contained but that there
will be notable differences with respect to my model specification and
special dataset. But maybe the code could serve as a template for my own
analysis and give me some ideas on how to solve the problems I am
facing.
Thank you for your assistance,
Karin.
-- 
*********************************************************************
                                                                      
  Karin Elsner             Institute of Agricultural Development      
                           in Central and Eastern Europe (IAMO)       
                           Emil-Abderhalden-Strasse 20                
                           06108 Halle                                
                           Germany                                    
                                                                      
                           --------------------------------------     
                                                                      
                           Phone +49 345 / 5 52 29 33                 
                           Fax   +49 345 / 5 00 81 77                 
                           Email elsner@iamo.uni-halle.de                
                    http://www.landw.uni-halle.de/iamo/iamo_e.htm     
                                                                      
*********************************************************************

---------- End of message ----------

From: "Eric Thode" <eric.thode@mail.uni-wuerzburg.de>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: RATS for estimation of AIDS using cross-section data
Date: Mon, 1 Feb 1999 11:01:28 +0200
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
X-mailer: Pegasus Mail for Win32 (v3.01d) (via Mercury MTS (Bindery) v1.40)

Hi Karin!

Concerning your problems estimating an AIDS consult the book of 
G. Hansen, "Quantitative Wirtschaftsforschung", Munich, 1993. 
(Sorry for quoting a German book, but since the question has been 
asked by a German this could be helpful for her)
On pages 307-315 you'll find some stuff about estimating an AIDS. 
Additionally, the book comes with a disk containing RATS 
examples of various topics the book deals with. I'm not 100 percent 
sure, but I think there's an example included on how to estimate an 
AIDS with RATS.

Hope this helps.

Eric
Dipl.-Vw. Eric Thode
Dept. of Economics (VWL 4)
University of Wuerzburg
Sanderring 2
97070 Wuerzburg

Tel.: +49-931-312928
Fax.: +49-931-312774

---------- End of message ----------

From: David Edgerton <david.edgerton@nek.lu.se>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: RE: RATS for estimation of AIDS using cross-section data
Date: Mon, 1 Feb 1999 12:39:37 +0100 
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2232.9) (via Mercury MTS (Bindery) v1.40)
Content-Type: multipart/mixed;

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_000_01BE4DD7.91F95EA0
Content-Type: text/plain

Hi!
I use the linearised form of AIDS in my econometric teaching. The following
exercise is given to my PhD students
 <<EX3_English.DOC>> 
This is the program I use to solve the  problem 
 <<EX3.PRG>> 
And here is another program for systemwise tests using Rao's test
 <<RAO_DYN.PRG>> 
Here is the data
 <<EX3.WKS>> 
Here is a TSP program for the NONLINEAR AIDS, which I've used in my
research. I've never used RATS to do this, but it should be quite easy to
adapt
 <<AID-ML.TSP>> 
You can also see my book "The  Econometrics of Demand Systems", published by
Kluwer 1996, which gives references to more data and programs (in TSP)
David
-----------------------------------------------------------------
David Edgerton
Dept of Economics 
P.O. Box 7082
S-220 07 Lund 
Sweden

tel: (+46) 46 222 8678
fax: (+46) 46 222 4613
e-mail: David.Edgerton@nek.lu.se <mailto:David.Edgerton@nek.lu.se> 


	-----Original Message-----
	From:	Karin Elsner [SMTP:elsner@iamo.uni-halle.de]
	Sent:	Monday, February 01, 1999 9:40 AM
	To:	RATS Discussion List
	Subject:	RATS for estimation of AIDS using cross-section data

	Dear colleagues,
	I am using RATS for doing cross-sectional analysis because this
software
	is available in the institute I am working at. More specifically, I
	estimate an Almost Ideal Demand System including socio-demographic
	variables.
	I face a lot of problems in transferring the algebraic model and the
	estimation procedures necessary for cross-section analysis (e.g.
	modified Heckamn procedure to account for zero expenditures, and so
on)
	into the RATS programme syntax. In order to avoid inventing the
wheel a
	hundredth time, I would like to ask whether somebody has estimated
an
	AIDS with RATS using cross-sectional or also time-series data and
	whether he or she could imagine giving me the respective programme
file.
	I know that this file would in no case be self-contained but that
there
	will be notable differences with respect to my model specification
and
	special dataset. But maybe the code could serve as a template for my
own
	analysis and give me some ideas on how to solve the problems I am
	facing.
	Thank you for your assistance,
	Karin.
	-- 
	
*********************************************************************
	

	  Karin Elsner             Institute of Agricultural Development

	                           in Central and Eastern Europe (IAMO)

	                           Emil-Abderhalden-Strasse 20

	                           06108 Halle

	                           Germany

	

	                           --------------------------------------

	

	                           Phone +49 345 / 5 52 29 33

	                           Fax   +49 345 / 5 00 81 77

	                           Email elsner@iamo.uni-halle.de

	                    http://www.landw.uni-halle.de/iamo/iamo_e.htm

	

	
*********************************************************************

------_=_NextPart_000_01BE4DD7.91F95EA0
Content-Type: application/msword;
	name="EX3_English.DOC"
Content-Transfer-Encoding: base64
Content-Disposition: attachment;
	filename="EX3_English.DOC"
Content-Location: ATT-4-136992E160B8D211A76B00805FC77430-E
	X3_EN%7E1.DOC
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------_=_NextPart_000_01BE4DD7.91F95EA0
Content-Type: application/octet-stream;
	name="EX3.PRG"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: attachment;
	filename="EX3.PRG"
Content-Location: ATT-0-0F6992E160B8D211A76B00805FC77430-E
	X3.PRG

CAL 1963
ALL 0 89:1
OPEN DATA EX3.WKS
DATA(ORG=3DOBS,FORMAT=3DWKS)

*
* FORM THE LAIDS SERIES
*


SET R1 =3D RMT+RFI+RDY
SET R2 =3D RFV+RPT+RBC
SET R3 =3D RNA+RSW+RBE
SET R4 =3D ROF+RSG+ROT
SET C1 =3D CMT+CFI+CDY
SET C2 =3D CFV+CPT+CBC
SET C3 =3D CNA+CSW+CBE
SET C4 =3D COF+CSG+COT
SET LP1 =3D LOG(C1/R1)
SET LP2 =3D LOG(C2/R2)
SET LP3 =3D LOG(C3/R3)
SET LP4 =3D LOG(C4/R4)
SET X =3D C1+C2+C3+C4
SET W1 =3D C1/X
SET W2 =3D C2/X
SET W3 =3D C3/X
SET W4 =3D C4/X
SET LP =3D LP1*W1 + LP2*W2 + LP3*W3 + LP4*W4
SET LX =3D LOG(X/POP) - LP

*
* FIT THE STATIC LAIDS MODEL
*

DISPLAY "STATIC LAIDS"
DISPLAY

LINREG(DEFINE=3DWEQ1) W1 / RES1
# CONSTANT LP1 LP2 LP3 LP4 LX

SET E1 =3D 1 + (%BETA(6)/W1)
SET E11 =3D (%BETA(2)-%BETA(6)*W1)/W1 - 1
SET E12 =3D (%BETA(3)-%BETA(6)*W2)/W1
SET E13 =3D (%BETA(4)-%BETA(6)*W3)/W1
SET E14 =3D (%BETA(5)-%BETA(6)*W4)/W1
SET EC11 =3D E11 + W1*E1
SET EC12 =3D E12 + W2*E1
SET EC13 =3D E13 + W3*E1
SET EC14 =3D E14 + W4*E1

LINREG(DEFINE=3DWEQ2) W2 / RES2
# CONSTANT LP1 LP2 LP3 LP4 LX

SET E2 =3D 1 + (%BETA(6)/W2)
SET E21 =3D (%BETA(2)-%BETA(6)*W1)/W2
SET E22 =3D (%BETA(3)-%BETA(6)*W2)/W2 - 1
SET E23 =3D (%BETA(4)-%BETA(6)*W3)/W2
SET E24 =3D (%BETA(5)-%BETA(6)*W4)/W2
SET EC21 =3D E21 + W1*E2
SET EC22 =3D E22 + W2*E2
SET EC23 =3D E23 + W3*E2
SET EC24 =3D E24 + W4*E2

LINREG(DEFINE=3DWEQ3) W3 / RES3
# CONSTANT LP1 LP2 LP3 LP4 LX

SET E3 =3D 1 + (%BETA(6)/W3)
SET E31 =3D (%BETA(2)-%BETA(6)*W1)/W3
SET E32 =3D (%BETA(3)-%BETA(6)*W2)/W3
SET E33 =3D (%BETA(4)-%BETA(6)*W3)/W3 - 1
SET E34 =3D (%BETA(5)-%BETA(6)*W4)/W3
SET EC31 =3D E31 + W1*E3
SET EC32 =3D E32 + W2*E3
SET EC33 =3D E33 + W3*E3
SET EC34 =3D E34 + W4*E3

LINREG(DEFINE=3DWEQ4) W4 / RES4
# CONSTANT LP1 LP2 LP3 LP4 LX

SET E4 =3D 1 + (%BETA(6)/W4)
SET E41 =3D (%BETA(2)-%BETA(6)*W1)/W4
SET E42 =3D (%BETA(3)-%BETA(6)*W2)/W4
SET E43 =3D (%BETA(4)-%BETA(6)*W3)/W4
SET E44 =3D (%BETA(5)-%BETA(6)*W4)/W4 - 1
SET EC41 =3D E41 + W1*E4
SET EC42 =3D E42 + W2*E4
SET EC43 =3D E43 + W3*E4
SET EC44 =3D E44 + W4*E4

COM NREGST =3D %NREG

TABLE / E1 E2 E3 E4
TABLE / E11 E22 E33 E44
TABLE / EC11 EC22 EC33 EC44

GRAPH(KEY=3DUPRIGHT) 4
 # E1
 # E2
 # E3
 # E4

GRAPH(KEY=3DUPRIGHT) 4
 # E11
 # E22
 # E33
 # E44

GRAPH(KEY=3DUPRIGHT) 4
 # EC11
 # EC22
 # EC33
 # EC44

*
* TEST FOR AUTOCORRELATION
*

DISPLAY
DISPLAY "EQUATIONWISE TESTS OF AUTOCORRELATION (FIRST ORDER B-G)"
DISPLAY

LINREG(NOPRINT) RES1
# CONSTANT LP1 LP2 LP3 LP4 LX RES1{1}
EXCLUDE
# RES1{1}

LINREG(NOPRINT) RES2
# CONSTANT LP1 LP2 LP3 LP4 LX RES2{1}
EXCLUDE
# RES2{1}

LINREG(NOPRINT) RES3
# CONSTANT LP1 LP2 LP3 LP4 LX RES3{1}
EXCLUDE
# RES3{1}

LINREG(NOPRINT) RES4
# CONSTANT LP1 LP2 LP3 LP4 LX RES4{1}
EXCLUDE
# RES4{1}


DISPLAY
DISPLAY "EQUATIONWISE TESTS OF AUTOCORRELATION (SECOND ORDER B-G)"
DISPLAY

LINREG(NOPRINT) RES1
# CONSTANT LP1 LP2 LP3 LP4 LX RES1{1 2}
EXCLUDE
# RES1{1 2}

LINREG(NOPRINT) RES2
# CONSTANT LP1 LP2 LP3 LP4 LX RES2{1 2}
EXCLUDE
# RES2{1 2}


LINREG(NOPRINT) RES3
# CONSTANT LP1 LP2 LP3 LP4 LX RES3{1 2}
EXCLUDE
# RES3{1 2}

LINREG(NOPRINT) RES4
# CONSTANT LP1 LP2 LP3 LP4 LX RES4{1 2}
EXCLUDE
# RES4{1 2}


DISPLAY
DISPLAY "SYSTEMWISE TEST OF AUTOCORRELATION (FIRST ORDER B-G)"
DISPLAY

COM NEQ=3D3
COM NREGS=3D9

EQUATION RESEQ1 RES1
# CONSTANT LP1 LP2 LP3 LP4 LX RES1{1} RES2{1} RES3{1}
EQUATION RESEQ2 RES2
# CONSTANT LP1 LP2 LP3 LP4 LX RES1{1} RES2{1} RES3{1}
EQUATION RESEQ3 RES3
# CONSTANT LP1 LP2 LP3 LP4 LX RES1{1} RES2{1} RES3{1}
SUR(NOPRINT) 3
# RESEQ1
# RESEQ2
# RESEQ3


TEST(ZEROS)
# 7 8 9 16 17 18 25 26 27
COM NREST =3D 9

COM CDW=3D%CDSTAT
DISPLAY
DISPLAY "LUTKEPOHL F-TEST"
DISPLAY
COM DF2 =3D %NOBS - NREGS
COM CWALD =3D (CDW*DF2)/FLOAT(NREST*%NOBS)
CDF FTEST CWALD NREST DF2

DISPLAY
DISPLAY "HOTELLING F-TEST"
DISPLAY
COM DFE =3D DF2-NEQ+1
COM CWALD =3D (CDW*DFE)/FLOAT(NREST*%NOBS)
CDF FTEST CWALD NREST DFE


*
* HABIT-FORMING LAIDS
*

DISPLAY
DISPLAY "DYNAMIC LAIDS"
DISPLAY

SMPL 64:1 89:1

LINREG(DEFINE=3DWEQL1) W1 / RESL1
# CONSTANT LP1 LP2 LP3 LP4 LX W1{1} W2{1} W3{1}

SET E1 =3D 1 + (%BETA(6)/W1)
SET E11 =3D (%BETA(2)-%BETA(6)*W1)/W1 - 1
SET E12 =3D (%BETA(3)-%BETA(6)*W2)/W1
SET E13 =3D (%BETA(4)-%BETA(6)*W3)/W1
SET E14 =3D (%BETA(5)-%BETA(6)*W4)/W1
SET EC11 =3D E11 + W1*E1
SET EC12 =3D E12 + W2*E1
SET EC13 =3D E13 + W3*E1
SET EC14 =3D E14 + W4*E1

LINREG(DEFINE=3DWEQL2) W2 / RESL2
# CONSTANT LP1 LP2 LP3 LP4 LX W1{1} W2{1} W3{1}

SET E2 =3D 1 + (%BETA(6)/W2)
SET E21 =3D (%BETA(2)-%BETA(6)*W1)/W2
SET E22 =3D (%BETA(3)-%BETA(6)*W2)/W2 - 1
SET E23 =3D (%BETA(4)-%BETA(6)*W3)/W2
SET E24 =3D (%BETA(5)-%BETA(6)*W4)/W2
SET EC21 =3D E21 + W1*E2
SET EC22 =3D E22 + W2*E2
SET EC23 =3D E23 + W3*E2
SET EC24 =3D E24 + W4*E2

LINREG(DEFINE=3DWEQL3) W3 / RESL3
# CONSTANT LP1 LP2 LP3 LP4 LX W1{1} W2{1} W3{1}

SET E3 =3D 1 + (%BETA(6)/W3)
SET E31 =3D (%BETA(2)-%BETA(6)*W1)/W3
SET E32 =3D (%BETA(3)-%BETA(6)*W2)/W3
SET E33 =3D (%BETA(4)-%BETA(6)*W3)/W3 - 1
SET E34 =3D (%BETA(5)-%BETA(6)*W4)/W3
SET EC31 =3D E31 + W1*E3
SET EC32 =3D E32 + W2*E3
SET EC33 =3D E33 + W3*E3
SET EC34 =3D E34 + W4*E3

LINREG(DEFINE=3DWEQL4) W4 / RESL4
# CONSTANT LP1 LP2 LP3 LP4 LX W1{1} W2{1} W3{1}

SET E4 =3D 1 + (%BETA(6)/W4)
SET E41 =3D (%BETA(2)-%BETA(6)*W1)/W4
SET E42 =3D (%BETA(3)-%BETA(6)*W2)/W4
SET E43 =3D (%BETA(4)-%BETA(6)*W3)/W4
SET E44 =3D (%BETA(5)-%BETA(6)*W4)/W4 - 1
SET EC41 =3D E41 + W1*E4
SET EC42 =3D E42 + W2*E4
SET EC43 =3D E43 + W3*E4
SET EC44 =3D E44 + W4*E4

COM NREGDY =3D %NREG

TABLE / E1 E2 E3 E4
TABLE / E11 E22 E33 E44
TABLE / EC11 EC22 EC33 EC44

GRAPH(KEY=3DUPRIGHT) 4
 # E1
 # E2
 # E3
 # E4

GRAPH(KEY=3DUPRIGHT) 4
 # E11
 # E22
 # E33
 # E44

GRAPH(KEY=3DUPRIGHT) 4
 # EC11
 # EC22
 # EC33
 # EC44

DISPLAY
DISPLAY "EQUATIONWISE FIRST ORDER B-G TEST"
DISPLAY

LINREG(NOPRINT) RESL1
# CONSTANT LP1 LP2 LP3 LP4 LX W1{1} W2{1} W3{1} RESL1{1}
EXCLUDE
# RESL1{1}

LINREG(NOPRINT) RESL2
# CONSTANT LP1 LP2 LP3 LP4 LX W1{1} W2{1} W3{1} RESL2{1}
EXCLUDE
# RESL2{1}

LINREG(NOPRINT) RESL3
# CONSTANT LP1 LP2 LP3 LP4 LX W1{1} W2{1} W3{1} RESL3{1}
EXCLUDE
# RESL3{1}

LINREG(NOPRINT) RESL4
# CONSTANT LP1 LP2 LP3 LP4 LX W1{1} W2{1} W3{1} RESL4{1}
EXCLUDE
# RESL4{1}

DISPLAY
DISPLAY "EQUATIONWISE SECOND ORDER B-G TEST"
DISPLAY

LINREG(NOPRINT) RESL1
# CONSTANT LP1 LP2 LP3 LP4 LX W1{1} W2{1} W3{1} RESL1{1} RESL1{2}
EXCLUDE
# RESL1{1} RESL1{2}

LINREG(NOPRINT) RESL2
# CONSTANT LP1 LP2 LP3 LP4 LX W1{1} W2{1} W3{1} RESL2{1} RESL2{2}
EXCLUDE
# RESL2{1} RESL2{2}

LINREG(NOPRINT) RESL3
# CONSTANT LP1 LP2 LP3 LP4 LX W1{1} W2{1} W3{1} RESL3{1} RESL3{2}
EXCLUDE
# RESL3{1} RESL3{2}

LINREG(NOPRINT) RESL4
# CONSTANT LP1 LP2 LP3 LP4 LX W1{1} W2{1} W3{1} RESL4{1} RESL4{2}
EXCLUDE
# RESL4{1} RESL4{2}

DISPLAY
DISPLAY "SYSTEMWISE FIRST ORDER B-G TEST"
DISPLAY

COM NEQ=3D3
COM NREGS=3D12

EQUATION RESLEQ1 RESL1
# CONSTANT LP1 LP2 LP3 LP4 LX W1{1} W2{1} W3{1} RESL1{1} RESL2{1} =
RESL3{1}
EQUATION RESLEQ2 RESL2
# CONSTANT LP1 LP2 LP3 LP4 LX W1{1} W2{1} W3{1} RESL1{1} RESL2{1} =
RESL3{1}
EQUATION RESLEQ3 RESL3
# CONSTANT LP1 LP2 LP3 LP4 LX W1{1} W2{1} W3{1} RESL1{1} RESL2{1} =
RESL3{1}
SUR(NOPRINT) 3
# RESLEQ1
# RESLEQ2
# RESLEQ3

TEST(ZEROS)
# 10 11 12 22 23 24 34 35 36
COM NREST =3D 9

COM CDW=3D%CDSTAT
DISPLAY
DISPLAY "LUTKEPOHL F-TEST"
DISPLAY
COM DF2 =3D %NOBS - NREGS
COM CWALD =3D (CDW*DF2)/FLOAT(NREST*%NOBS)
CDF FTEST CWALD NREST DF2

DISPLAY
DISPLAY "HOTELLING F-TEST"
DISPLAY
COM DFE =3D DF2-NEQ+1
COM CWALD =3D (CDW*DFE)/FLOAT(NREST*%NOBS)
CDF FTEST CWALD NREST DFE


*
* TEST OF HOMOGENEITY AND SYMMETRY
*

SMPL 63:1 89:1

SUR(NOPRINT) 3
# WEQ1
# WEQ2
# WEQ3

DISPLAY "STATIC TEST OF HOMOGENEITY"
RESTRICT 3
# 2 3 4 5
# 1 1 1 1 0
# 8 9 10 11
# 1 1 1 1 0
# 14 15 16 17
# 1 1 1 1 0
COM  NREST =3D 3

COM CDW=3D%CDSTAT
DISPLAY
DISPLAY "LUTKEPOHL F-TEST"
DISPLAY
COM DF2 =3D %NOBS - NREGST
COM CWALD =3D (CDW*DF2)/FLOAT(NREST*%NOBS)
CDF FTEST CWALD NREST DF2

DISPLAY
DISPLAY "HOTELLING F-TEST"
DISPLAY
COM DFE =3D DF2-NEQ+1
COM CWALD =3D (CDW*DFE)/FLOAT(NREST*%NOBS)
CDF FTEST CWALD NREST DFE


DISPLAY
DISPLAY "STATIC TEST OF SYMMETRY"
RESTRICT 6
# 2 3 4 5
# 1 1 1 1 0
# 8 9 10 11
# 1 1 1 1 0
# 14 15 16 17
# 1 1 1 1 0
# 3 8
# 1 -1 0
# 4 14
# 1 -1 0
# 10 15
# 1 -1 0
COM  NREST =3D 6

COM CDW=3D%CDSTAT
DISPLAY
DISPLAY "LUTKEPOHL F-TEST"
DISPLAY
COM DF2 =3D %NOBS - NREGST
COM CWALD =3D (CDW*DF2)/FLOAT(NREST*%NOBS)
CDF FTEST CWALD NREST DF2

DISPLAY
DISPLAY "HOTELLING F-TEST"
DISPLAY
COM DFE =3D DF2-NEQ+1
COM CWALD =3D (CDW*DFE)/FLOAT(NREST*%NOBS)
CDF FTEST CWALD NREST DFE


SMPL 64:1 89:1

SUR(NOPRINT) 3
# WEQL1
# WEQL2
# WEQL3

DISPLAY "TEST OF DYNAMICS"
TEST(ZEROS)
 # 7 8 9 16 17 18 25 26 27
COM  NREST =3D 9

COM CDW=3D%CDSTAT
DISPLAY
DISPLAY "LUTKEPOHL F-TEST"
DISPLAY
COM DF2 =3D %NOBS - NREGDY
COM CWALD =3D (CDW*DF2)/FLOAT(NREST*%NOBS)
CDF FTEST CWALD NREST DF2

DISPLAY
DISPLAY "HOTELLING F-TEST"
DISPLAY
COM DFE =3D DF2-NEQ+1
COM CWALD =3D (CDW*DFE)/FLOAT(NREST*%NOBS)
CDF FTEST CWALD NREST DFE


DISP
DISPLAY "DYNAMIC TEST OF HOMOGENEITY"
RESTRICT 3
# 2 3 4 5
# 1 1 1 1 0
# 11 12 13 14
# 1 1 1 1 0
# 20 21 22 23
# 1 1 1 1 0
COM  NREST =3D 3

COM CDW=3D%CDSTAT
DISPLAY
DISPLAY "LUTKEPOHL F-TEST"
DISPLAY
COM DF2 =3D %NOBS - NREGDY
COM CWALD =3D (CDW*DF2)/FLOAT(NREST*%NOBS)
CDF FTEST CWALD NREST DF2

DISPLAY
DISPLAY "HOTELLING F-TEST"
DISPLAY
COM DFE =3D DF2-NEQ+1
COM CWALD =3D (CDW*DFE)/FLOAT(NREST*%NOBS)
CDF FTEST CWALD NREST DFE


DISPLAY
DISPLAY "DYNAMIC TEST OF SYMMETRY"
RESTRICT 6
# 2 3 4 5
# 1 1 1 1 0
# 11 12 13 14
# 1 1 1 1 0
# 20 21 22 23
# 1 1 1 1 0
# 3 11
# 1 -1 0
# 4 20
# 1 -1 0
# 13 21
# 1 -1 0
COM  NREST =3D 6

COM CDW=3D%CDSTAT
DISPLAY
DISPLAY "LUTKEPOHL F-TEST"
DISPLAY
COM DF2 =3D %NOBS - NREGDY
COM CWALD =3D (CDW*DF2)/FLOAT(NREST*%NOBS)
CDF FTEST CWALD NREST DF2

DISPLAY
DISPLAY "HOTELLING F-TEST"
DISPLAY
COM DFE =3D DF2-NEQ+1
COM CWALD =3D (CDW*DFE)/FLOAT(NREST*%NOBS)
CDF FTEST CWALD NREST DFE





------_=_NextPart_000_01BE4DD7.91F95EA0
Content-Type: application/octet-stream;
	name="RAO_DYN.PRG"
Content-Disposition: attachment;
	filename="RAO_DYN.PRG"
Content-Location: ATT-1-106992E160B8D211A76B00805FC77430-R
	AO_DYN.PRG

CAL 1963
ALL 0 89:1
OPEN DATA EX3.WKS
DATA(ORG=OBS,FORMAT=WKS)

*
* FORM THE LAIDS SERIES
*

SET R1 = RMT+RFI+RDY
SET R2 = RFV+RPT+RBC
SET R3 = RNA+RSW+RBE
SET R4 = ROF+RSG+ROT
SET C1 = CMT+CFI+CDY
SET C2 = CFV+CPT+CBC
SET C3 = CNA+CSW+CBE
SET C4 = COF+CSG+COT
SET LP1 = LOG(C1/R1)
SET LP2 = LOG(C2/R2)
SET LP3 = LOG(C3/R3)
SET LP4 = LOG(C4/R4)
SET X = C1+C2+C3+C4
SET W1 = C1/X
SET W2 = C2/X
SET W3 = C3/X
SET W4 = C4/X
SET LP = LP1*W1 + LP2*W2 + LP3*W3 + LP4*W4
SET LX = LOG(X/POP) - LP

*
* FIT THE DYNAMIC LAIDS MODEL
*

DISPLAY "DYNAMIC LAIDS"
DISPLAY

LINREG(DEFINE=WEQ1,NOPRINT) W1 / RES1
# CONSTANT LP1 LP2 LP3 LP4 LX W1{1} W2{1} W3{1}

LINREG(DEFINE=WEQ2,NOPRINT) W2 / RES2
# CONSTANT LP1 LP2 LP3 LP4 LX W1{1} W2{1} W3{1}

LINREG(DEFINE=WEQ3,NOPRINT) W3 / RES3
# CONSTANT LP1 LP2 LP3 LP4 LX W1{1} W2{1} W3{1}


COM NREGS=%NREG
COM NEQ = 3
LIST STATSUR = WEQ1 WEQ2 WEQ3
SUR(NOPRINT,OUTSIGMA=SIGU) NEQ
CARDS STATSUR
COM LLU=LOG(%DET(SIGU))

DISPLAY "WALD TEST OF HOMOGENEITY"
DISPLAY
DISPLAY "ASYMPTOTIC TEST"
DISPLAY

COM NRESTW=3

RESTRICT NRESTW
# 2 3 4 5
# 1 1 1 1 0
# 11 12 13 14
# 1 1 1 1 0
# 20 21 22 23
# 1 1 1 1 0

DISPLAY
DISPLAY "APPROX F-TEST"
DISPLAY

COM CDW = %CDSTAT
COM DF2 = %NOBS - NREGS
COM CWALD = (CDW*DF2)/FLOAT(NRESTW*%NOBS)
CDF FTEST CWALD NRESTW DF2

DISPLAY
DISPLAY "HOTELLING F-TEST"
DISPLAY

COM DFE = DF2-NEQ+1
COM CWALD = (CDW*DFE)/FLOAT(NRESTW*%NOBS)
CDF FTEST CWALD NRESTW DFE


DISPLAY
DISPLAY "LRT OF HOMOGENEITY"
RESTRICT(NOPRINT,REPLACE) NRESTW
# 2 3 4 5
# 1 1 1 1 0
# 11 12 13 14
# 1 1 1 1 0
# 20 21 22 23
# 1 1 1 1 0


SUR(CREATE,NOPRINT,OUTSIGMA=SIGR,ITER=20) NEQ
CARDS STATSUR
COM LLR=LOG(%DET(SIGR))

COM LRT = %NOBS*(LLR-LLU)
CDF CHISQUARED LRT NRESTW

COM P=NRESTW
COM R=(P/2.)-1
COM S=SQRT( ((P**2) - 4) / ((NEQ**2) + ((NRESTW/FLOAT(NEQ))**2) - 5) )
COM H=DF2-(NEQ-(NRESTW/FLOAT(NEQ))+1)/2.
COM Q=H*S-R
COM US=EXP((LLR-LLU))**(1./S)
COM RAO= (Q/P)*(US-1)

COM QL=FIX(Q); COM QU=QL+1
COM PVL=%FTEST(RAO,P,QL)
COM PVU=%FTEST(RAO,P,QU)
COM PRAO = (Q-QL)*PVU + (QU-Q)*PVL
DISPLAY "RAO TEST =" RAO " WITH DF" P " , " Q
DISPLAY "P-VALUE =" PRAO


SUR(NOPRINT,OUTSIGMA=SIGU) NEQ
CARDS STATSUR

DISPLAY "WALD TEST OF SYMMETRY"
DISPLAY
DISPLAY "ASYMPTOTIC TEST"
DISPLAY

COM NRESTS=6

RESTRICT NRESTS
# 2 3 4 5
# 1 1 1 1 0
# 11 12 13 14
# 1 1 1 1 0
# 20 21 22 23
# 1 1 1 1 0
# 3 11
# 1 -1 0
# 4 20
# 1 -1 0
# 13 21
# 1 -1 0

DISPLAY
DISPLAY "APPROX F-TEST"
DISPLAY

COM CDW = %CDSTAT
COM DF2 = %NOBS - NREGS
COM CWALD = (CDW*DF2)/FLOAT(NRESTS*%NOBS)
CDF FTEST CWALD NRESTS DF2

DISPLAY
DISPLAY "HOTELLING F-TEST"
DISPLAY

COM CWALD = (CDW*DFE)/FLOAT(NRESTS*%NOBS)
CDF FTEST CWALD NRESTS DFE

DISPLAY
DISPLAY "LRT OF SYMMETRY"
RESTRICT(NOPRINT,REPLACE) NRESTS
# 2 3 4 5
# 1 1 1 1 0
# 11 12 13 14
# 1 1 1 1 0
# 20 21 22 23
# 1 1 1 1 0
# 3 11
# 1 -1 0
# 4 20
# 1 -1 0
# 13 21
# 1 -1 0

SUR(CREATE,NOPRINT,OUTSIGMA=SIGR,ITER=20) NEQ
CARDS STATSUR
COM LLR=LOG(%DET(SIGR))

COM LRT = %NOBS*(LLR-LLU)
CDF CHISQUARED LRT NRESTS

COM P=NRESTS
COM R=(P/2.)-1
COM S=SQRT( ((P**2) - 4) / ((NEQ**2) + ((NRESTS/FLOAT(NEQ))**2) - 5) )
COM H=DF2-(NEQ-(NRESTS/FLOAT(NEQ))+1)/2.
COM Q=H*S-R
COM US=EXP((LLR-LLU))**(1./S)
COM RAO= (Q/P)*(US-1)

COM QL=FIX(Q); COM QU=QL+1
COM PVL=%FTEST(RAO,P,QL)
COM PVU=%FTEST(RAO,P,QU)
COM PRAO = (Q-QL)*PVU + (QU-Q)*PVL
DISP
DISPLAY "RAO TEST =" RAO " WITH DF" P " , " Q
DISPLAY "P-VALUE =" PRAO






------_=_NextPart_000_01BE4DD7.91F95EA0
Content-Type: application/octet-stream;
	name="EX3.WKS"
Content-Transfer-Encoding: base64
Content-Disposition: attachment;
	filename="EX3.WKS"
Content-Location: ATT-2-116992E160B8D211A76B00805FC77430-E
	X3.WKS
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------_=_NextPart_000_01BE4DD7.91F95EA0
Content-Type: application/octet-stream;
	name="AID-ML.TSP"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: attachment;
	filename="AID-ML.TSP"
Content-Location: ATT-3-126992E160B8D211A76B00805FC77430-A
	ID-ML.TSP

?
? ROBUST DYNAMIC AIDS -  S.E.ELAS         AID-D.TSP

?   A0 FIXED
?   HOMOGENEITY AND SYMMETRY CONDITIONS CAN BE APPLIED
?   LR, WALD AND SSR TESTS=20
?   ELASTICITIES WITH THEIR STANDARD ERRORS
?   PREDICTIONS WITH THEIR STANDARD ERRORS

? NONINDENTED LINES, WITH *** AT END OF ROW, MUST BE CHANGED
?   FOR DIFFERENT RUNS, DEPENDING ON NUMBER OF VARIABLES
? COMMANDS OVER SEVERAL LINES WITH !** AT END MUST BE
?   CHANGED IN SOME LINE
? COMMANDS WITH //* AT END VARY IN NUMBER DEPENDING ON
? EQUATIONS ETC ARE NUBERED 1, 2, ...=20

?   N =3D SIZE OF SYSTEM
?   MIT =3D MAXIMUM NUMBER OF ITERATIONS

?   for . =3D D (GENERAL DYNAMIC), T (STATIC)
?   MET. =3D 1 USE MODEL .
?   if met. =3D 1 then
?   HOMSYM. =3D 1 HOMSYM RESTRICTIONS FOR MODEL .
?   IELAS. =3D 1 ELASTICITES FOR MODEL .
?   ISEELAS. =3D 0 (NONE), 1 (MEAN), 2 (MEAN+LAST), 3 (ALL), MODL .
?   WSTORE. =3D 1 STORE ESTIMATED BUDGET SHARES, MODEL .
?   ISTORE. =3D 1 STORE MODEL .
?   ISTHS. =3D 1 STORE HOMSYM FOR MODEL .
?   IDTEST. =3D 1 DIAGNOSTIC TESTS FOR MODEL .
?   IDSYS. =3D 1 SYSTEM DIAGNOSTIC TESTS
?   IPRED. =3D 1 PREDICTIONS FOR MODEL .
?   ISEPRED. =3D 0 (NONE), 1 (PRED), 2 (PRED+SMPL), FOR MODEL .

?   IRTEST =3D RESTRICTION TESTS: 0 (NONE), 1 (H/S/D), 2 (ALL)
?   IELASHS =3D 1 ELASTICITIES FOR HOMSYM (if also for model)
?   ISELASHS =3D 1 SAME S.E. ELAS AS MODEL
?   IPREDHS =3D 1 PREDICTIONS FOR HOMSYM (if also for model)
?   ISPREDHS =3D 1 SAME S.E. PRED AS MODEL
?   IDTESTHS =3D 1 DIAGNOSTIC TESTS FOR HOMSYM (if also for model)
?   IDSYSHS =3D 1 SYSTEM DIAGNOSTICS FOR HOMSYM (if also for model)
?   IROBUST =3D 1 HETEROSCEDASTICITY ROBUST STANDARD ERRORS
?   IPRINT =3D 0 (NORMAL), 1 (INCREASED), 2 (MAXIMUM)=20

? PARAMETER A0 IS SET EQUAL TO LOG(SUBSISTENCE TOT CONSUMP)



? SET DATA PERIOD AND SAMPLE PERIOD

freq a;                       ?***;

set td1 =3D 1963;                   ?***;
set tdt =3D 1989;                   ?***;
set te1 =3D 1963;                   ?***;
set tet =3D 1989;                   ?***;

	smpl td1,tdt;
	rename @smpl datsmpl2;
	supres @smpl;

	in '\livsmed\data\tsp\foodbnk2';


? STATE NUMBER OF EQUATIONS, BASIC LISTS AND VARIABLES TO
? BE USED. SET A0

set n =3D 3;                      ?***;
set mit =3D 80;                     ?***;

set metD =3D 1;                     ?***;
set homsymD =3D 0;                    ?***;
set ielasD =3D 0;                   ?***;
set iseelasD =3D 0;                   ?***;
set ipredD =3D 0;                   ?***;
set isepredD =3D 0;                   ?***;
set istoreD =3D 0;                    ?***;
set isthsD =3D 0;                   ?***;
set wstoreD =3D 0;                    ?***;
set idtestD =3D 0;                    ?***;
set idsysD =3D 0;                   ?***;

set metT =3D 0;                     ?***;
set homsymT =3D 0;                    ?***;
set ielasT =3D 0;                   ?***;
set iseelasT =3D 0;                   ?***;
set ipredT =3D 0;                   ?***;
set isepredT =3D 0;                   ?***;
set istoreT =3D 0;                    ?***;
set isthsT =3D 0;                   ?***;
set wstoreT =3D 0;                    ?***;
set idtestT =3D 0;                    ?***;
set idsysT =3D 0;                   ?***;

set irtest =3D 1;                   ?***;
set iprint =3D 0;                   ?***;
set irobust =3D 1;                    ?***;
set critval =3D 2;                    ?***;
set ielashs =3D 0;                    ?***;
set iselashs =3D 0;                   ?***;
set ipredhs =3D 0;                    ?***;
set ispredhs =3D 0;                   ?***;
set idtesths =3D 0;                   ?***;
set idsyshs =3D 0;                    ?***;

list dlist 1-3;                   ?***;
list dlist2 2-3;                    ?//*;
list dlist3 3;                      ?//*;
list dgnl 11 22 33;                   ?***;
list symlist ws32;                    ?***;
list lsymlist ls32;                   ?***;
list dynlist wd22-wd23 wd32-wd33;             ?***;
	list dlist1 dlist;

npop=3Dmptotal;                     ?***;
c1=3D cmeat;                      ?***;
c2=3D cfish;                      ?***;
c3=3D cdairy;                     ?***;
r1=3D rmeat;                      ?***;
r2=3D rfish;                      ?***;
r3=3D rdairy;                     ?***;
set a0 =3D log(1.2);                    ?***;

	set n1 =3D n - 1;
	set dfex1 =3D n1;

? CALCULATE LOG(PRICES), LOG(TOT CONSUMP) AND BUDGET SHARES

	if metD^=3D0; then; goto 116;
	set td0 =3D td1-1; set t0 =3D te1-1;
	smpl td0,tdt;
	lx=3D0;
	dot dlist;
		wl.=3D0; w.=3D0; lp.=3D0;
	enddot;
	smpl td1,tdt;
	goto 115;
116 set td0 =3D td1; set t0 =3D te1;
115 set td1 =3D td0+1; set t1 =3D t0+1; set t2 =3D t1+1;
	set t3 =3D t1+2; set tt =3D tet; set tt1 =3D tt+1;
	set datsmpl2(1) =3D td0;
	dot datsmpl smplest smplest1 smplest2 smplpred smplall smplall2;
		copy datsmpl2 .;
	enddot;
	set datsmpl(1) =3D td1;
	set smplest(1) =3D t1; set smplest(2) =3D tt;
	set smplest2(1) =3D t0; set smplest2(2) =3D tt;
	if tt>=3Dtdt; then; goto 710;
	set smplpred(1) =3D tt1;
	set smplall(1) =3D t1; set smplall2(1) =3D t0;
	goto 711;
710 copy smplest smplall; copy smplest2 smplall2;

711 set atmean =3D 1; set samplper=3D1; set predper=3D1;
	cx=3D0;
	dot dlist;
		pr.=3Dc./r.;
		lp. =3D log(pr.);
		cx =3D cx+c.;
	enddot;
	x =3D cx/npop;
	lx =3Dlog(x);
	dot dlist;
		w. =3D c./cx;
		cp. =3D c./npop;
	enddot;
	totexp =3D cnondur+cserv+cfoodhom+crestcf;
	linc =3D log(cdspinc/npop);
	lexp =3D log(totexp/npop);

	smpl datsmpl;
	dot dlist;
		wl. =3D w.(-1);
	enddot;

	smpl smplall;


? CALCULATE COMPLEMENTARY LISTS ETC

	list eqlist eq2;
	list wlist w2;
	list plist pr1 pr2;
	list homlist wh2;
	list lhomlist lh2;
	list thlist pth11;
	list vlists c lp1 lp2;
	list vlistd vlists wl2;
	list addlist pa1 pb1 pg11 pg12;
	list bgs1flt bgs1f2;
	list bgs1dlt bgs1d2;
	list bgs2dlt bgs2d2;
	list bps1flt bps1f1 bps1f2;
	list bps1dlt bps1d1 bps1d2;
	list bps2dlt bps2d1 bps2d2;
	list achs1flt achs1f1 achs1f2;
	list achs1dlt achs1d1 achs1d2;
	list achs2dlt achs2d1 achs2d2;
	list rsts2flt rsts2f2;
	list rsts2dlt rsts2d2;
	list rsts3dlt rsts3d2;
	list hwutcslt hwutcs2;
	list hwudislt hwudis2;
	list eqult equ2;
	list equ2lt equ21 equ22;
	list eqwlt eqw2;
	if n.lt.3; then; goto 400;

	dot dlist3;
		list eqlist eqlist eq.;
		list wlist wlist w.;
		list plist plist pr.;
		list homlist homlist wh.;
		list lhomlist lhomlist lh.;
		list vlists vlists lp.;
		list vlistd vlistd lp. wl.;
		list bgs1flt bgs1flt bgs1f.;
		list bgs1dlt bgs1dlt bgs1d.;
		list bgs2dlt bgs2dlt bgs2d.;
		list bps1flt bps1flt bps1f.;
		list bps1dlt bps1dlt bps1d.;
		list bps2dlt bps2dlt bps2d.;
		list achs1flt achs1flt achs1f.;
		list achs1dlt achs1dlt achs1d.;
		list achs2dlt achs2dlt achs2d.;
		list rsts2flt rsts2flt rsts2f.;
		list rsts2dlt rsts2dlt rsts2d.;
		list rsts3dlt rsts3dlt rsts3d.;
		list hwutcslt hwutcslt hwutcs.;
		list hwudislt hwudislt hwudis.;
		list eqult eqult equ.;
		list equ2lt equ2lt equ2.;
		list eqwlt eqwlt eqw.;
		if n.gt.9; then; list addlist addlist pg01.;
			else; list addlist addlist pg1.;
	enddot;

400 dot dlist2;
		list thlist thlist pth1. pth.1;
	enddot;
	list lstl11 BGS1F BGS1D;
	list lstl12 BGS2D;
	list lstl30 BPS1F BPS1D BPS2D;
	list lstl31 ACHS1F ACHS1D;
	list lstl32 ACHS2D;
	list lstlst1 BGS1F BGS1D BGS2D;
	list lstlst2 RSTS2F RSTS2D RSTS3D HWUTCS HWUDIS;
	list lstlst3 BPS1F BPS1D BPS2D ACHS1F ACHS1D ACHS2D;
	list syslt lstlst1 lstlst3 lstlst2;
	dot ep sp epc spc;
		dot dlist;
list ..lst ..1-..3;                   ?***;
		enddot;
	enddot;

	list parlst a0;
	list mlist logl rsqdet arsqdet rsqtrw arsqtrw akaike;
	dot dlist;
		list mlist mlist rsq. arsq. dw.;
		list parlst parlst a. b.;
		dot dlist;
			list parlst parlst g.. th..;
		enddot;
	enddot;
	dot mlist;
		set . =3D 0;
	enddot;
	list parlst parlst mlist;

	dot rsq ar2;
		dot dlist;
			set .DcT. =3D 0;
		enddot;
	enddot;
	dot r2det ar2dt r2trw ar2tr;
		set .DcT =3D 0;
	enddot;

	if irtest =3D 0; then; goto 917;
	list tlistT H S;
	list tlistD H S D;
	list tlistH1 ScH;
	list tlistH2 DcH;
	list tlistS2 DcS;
	list Hlt homlist;
	list Slt homlist symlist;
	list Dlt dynlist;
	list ScHlt symlist;
	list DcHlt Dlt; list DcSlt Dlt;

	if irtest =3D 1; then; goto 916;
	list tlistT tlistT ;
	list tlistD tlistD DH DS ;
	list tlistH2 tlistH2 DScH;
	list DHlt dynlist homlist;
	list DSlt dynlist homlist symlist;
	list DScHlt dynlist symlist;

916 list wtest tlistH1 tlistH2 tlistS2 tlistD;
	dot wtest;
		list nam. .;
	enddot;
	dot wld pwd cwd pcw;
		dot wtest;
			set ..=3D0;
		enddot;
	enddot;
	set exwldH=3D0; set pxwldH=3D0;

917 dot D T H S;
		set logl.=3D0; set p.=3D0;
		list nam. .;
	enddot;
	set iconvT=3D0;

	compress;

	if istoreT=3D0&isthsT=3D0&istoreD=3D0&isthsD=3D0;
	then; goto 966;
	dot bmep bmepc bmvp bmvpc lmep lmepc lmvp lmvpc;
		mform(nrow=3Dn,ncol=3Dn) . =3D 0;
	enddot;
	dot bve bvv lve lvv;
		mform(nrow=3D1,ncol=3Dn) . =3D 0;
	enddot;
	dot w cp c;
		dot dlist;
			set pr2.t.=3D0; set .trms.=3D0;
		enddot;
	enddot;
	smpl smplall2;
	list varlst w1 c1 cp1;
	list predlst wt1 ct1 pct1;
	list spredlst swt1 sct1 spct1;
	list elaslt e1 ep1lst epc1lst;
	list selaslt se1 sp1lst spc1lst;
	dot dlist2;
		list varlst varlst w. c. cp.;
		list predlst predlst wt. ct. pct.;
		list spredlst spredlst swt. sct. spct.;
		list elaslt elaslt e. ep.lst epc.lst;
		list selaslt selaslt se. sp.lst spc.lst;
	enddot;
	dot elaslt selaslt predlst spredlst;
		.=3D0;
	enddot;
	smpl smplall;
	list lrtlist lrt clrt plrt pclrt exlrt pxlrt;
	dot lrtlist;
		dot H S ScH D;
			set ..=3D0;
		enddot;
	enddot;
	dot syslt;
		set ex.=3D0; set px.=3D0; set lr.=3D0; set pl.=3D0;
	enddot;
966 list dtlist BG1 BG2 BP1 BP2 ARCH1 ARCH2 RSET2 RSET3 HWUTC HWUDI;

	dot dlist;
		list namBG1. BG1.;
		list namBG2. BG2.;
		list namBP1. BP1.;
		list namBP2. BP2.;
		list namAR1. ARCH1.;
		list namAR2. ARCH2.;
		list namRS2. RSET2.;
		list namRS3. RSET3.;
		list namHTC. HWUTC.;
		list namHDI. HWUDI.;
	enddot;

	list BG1lt uhat(-1);
	list BG2lt uhat(-1) uhat(-2);
	list BP1lt what;
	list BP2lt what what2;
	list ARCH1lt uhat2(-1);
	list ARCH2lt uhat2(-1) uhat2(-2);
	list RSET2lt what2;
	list RSET3lt what2 what3;
	list HWUTClt lxexph;
	list HWUDIlt lxinch;
	list clist constant;

	set mcsmsq=3D0;
	dot dtlist;
		dot dlist;
			set f..=3D0; set p..=3D0;
		enddot;
	enddot;
	dot BJ BJAS PBJAS CUSMSQ;
		dot dlist;
			set ..=3D0;
		enddot;
	enddot;

	smpl smplest;
	rename @nob nobs;
	set ntot =3D nobs*n1;
	dot w cp c;
		dot dlist;
			msd(noprint) ..;
			set v.. =3D (nobs-1)*@var(1)/nobs;
			set .lt. =3D @mean(1);
		enddot;
	enddot;
	msd(noprint) x plist;
	smpl smplest2;
	set lx(t0) =3D log(@mean(1));
	set inum =3D 2;
	dot dlist;
		set lp.(t0) =3D log(@mean(inum));
		set wl.(t0) =3D wlt.;
		set inum =3D inum + 1;
	enddot;
	smpl smplest;

? SPECIFY PARAMETERS

	dot dlist;
		param a. b. dp.0 dp.1 dp.2 dp.3;
		dot dlist;
			param g..;
		enddot;
	enddot;

	compress;


? SPECIFY EQUATIONS RESTRICTIONS

frml pth11, th11 =3D - th21 - th31;             ?***;
	dot dlist2;
frml pth1., th1. =3D - th2. - th3.;             ?***;
frml pth.1, th.1 =3D - th.2 - th.3;             ?***;
frml bgs1f., uht. =3D wwt. + dp.2*uht2(-1) + dp.3*uht3(-1);   ?***;
frml rsts2f., w. =3D wwt. + dp.1*wht21 + dp.2*wht22 + dp.3*wht23; ?***;

		frml eq., w. =3D wwt.;
		copy eq. eqw.;
		frml equ., uht. =3D wwt.;
		frml bgs1d., uht. =3D wwt. + dp.1*uht.(-1);
		frml bgs2d., uht. =3D wwt. + dp.1*uht.(-1) + dp.2*uht.(-2);
		frml rsts2d., w. =3D wwt. + dp.2*wht2.;
		frml rsts3d., w. =3D wwt. + dp.2*wht2. + dp.3*wht3.;
		frml hwutcs., w. =3D wwt. + dp.1*lxexph;
		frml hwudis., w. =3D wwt. + dp.1*lxinch;
		eqsub pth11 pth.1;
	enddot;
	dot dlist;
frml bps1f., uht2. =3D dp.0 + dp.2*wht2 + dp.3*wht3;      ?***;
frml achs1f., uht2. =3D dp.0 + dp.1*uht21(-1) + dp.2*uht22(-1) +
			 dp.3*uht23(-1);              ?//*;
		frml equ2., uht2. =3D dp.0;
		frml bps1d., uht2. =3D dp.0 + dp.1*wht.;
		frml bps2d., uht2. =3D dp.0 + dp.1*wht. + dp.2*wht2.;
		frml achs1d., uht2. =3D dp.0 + dp.1*uht2.(-1);
		frml achs2d., uht2. =3D dp.0 + dp.1*uht2.(-1) + dp.2*uht2.(-2);
	enddot;

frml pa1, a1 =3D 1 - a2 - a3;               ?***;
frml pb1, b1 =3D - b2 - b3;               ?***;
frml cpp, pp =3D a0 + pp1 + pp2 + pp3;            ?***;

	dot dlist;

frml pg1., g1. =3D - g2. - g3.;               ?***;
frml xp., a. + th.1*wl1 + th.2*wl2 + th.3*wl3
 + 0.5 * (g.1*lp1 + g.2*lp2 + g.3*lp3);           ?!**;
frml cwt., wwt. =3D a.+ b.*(lx-pp)
+ th.1*wl1 + th.2*wl2 + th.3*wl3
+ g.1*lp1 + g.2*lp2 + g.3*lp3;              ?!**;
frml xx., xp. + 0.5 * (g1.*lp1 + g2.*lp2 + g3.*lp3);      ?!**;
frml cep1., eep1. =3D (g1. - b1* xx.)/wwt1;     ?//*;
frml cep2., eep2. =3D (g2. - b2* xx.)/wwt2;     ?//*;
frml cep3., eep3. =3D (g3. - b3* xx.)/wwt3;     ?//*;
frml cpc1., eepc1. =3D eep1. + ee1*wwt.;            ?//*;
frml cpc2., eepc2. =3D eep2. + ee2*wwt.;            ?//*;
frml cpc3., eepc3. =3D eep3. + ee3*wwt.;            ?//*;
		frml ce., ee. =3D 1 + (b./wwt.);
		frml pp., lp. * xp.;
	enddot;

	dot dlist;
		eqsub xp. addlist thlist;
		eqsub xx. xp. addlist;
		eqsub pp. xp.;
		eqsub cpp pp.;
	enddot;
	frml dpp, pp =3D phat;
	dot dlist;
		eqsub cwt. addlist thlist;
		copy cwt. dwt.;
		eqsub cwt. cpp;
		eqsub dwt. dpp;
	enddot;
	dot lstlst1 lstlst2 equ eqw;
		dot dlist2;
			eqsub .. dwt.;
		enddot;
	enddot;
	dot dlist2;
frml wh., g.1 + g.2 + g.3;                ?***;
		eqsub eq. cwt.;
		dot dlist2;
			frml wd.., th..;
		enddot;
	enddot;
dot dlist3;                     ?//*;
frml ws.2, g.2 - g2.;                 ?//*;
enddot;                       ?//*;
	list alist addlist;

	mat identn =3D ident(n);
	print smplest;
	if tet<tdt; then; print smplpred;
	msd(noprint,cova) wlist;
	mat detw =3D det(@cova);
	rename @cova sigw;

?   STATIC WITH (NOBS - 1) OBSERVATIONS (NOBS OBS IF ONLY MODEL)

	if metT=3D0; then; goto 480;

	dot dlist;
		dot dlist;
			const th.. 0;
		enddot;
	enddot;

	if irobust =3D 1; then; lsq(maxit=3Dmit,robust) eqlist; else;
		lsq(maxit=3Dmit) eqlist;
	set iconvT =3D @ifconv; if iconvT^=3D1; then; goto 480;
	analyz(noprint) addlist ;

	rescalc namT tlistT ielasT iseelasT ipredT isepredT istoreT=20
		idtestT idsysT 1 0 sigmaT invsigT coef ses coefa sesa;
	dot dlist;
		set rsqT. =3D rsq.;
		set arsqT. =3D arsq.;
	enddot;
	set rsqdetT =3D rsqdet; set arsqdetT =3D arsqdet;
	set rsqtrwT =3D rsqtrw; set arsqtrwT =3D arsqtrw;

	if wstoreT=3D0; then; goto 885;
	out '\livsmed\data\tsp\foodbnk2';
	dot dlist;
genr cwt. nw1.usTm;                   ?***;
	enddot;
	out;
885 if istoreT=3D0; then; goto 886;
out '\livsmed\data\aids\3eqn\ml\m\T-a';           ?***;
	store1 ielasT iseelasT ipredT isepredT idtestT idsysT tlistT 1 0=20
		coef ses coefa sesa;
	out;

886 if homsymT^=3D0; then; homsym namT ielasT iseelasT ipredT isepredT=20
		isthsT istoreT idtestT idsysT 1;


? GENERAL DYNAMIC ESTIMATES

480 if metD=3D0; then; goto 999;

	dot dlist;
		dot dlist;
			param th..;
		enddot;
	enddot;

	if irobust =3D 1; then; lsq(maxit=3Dmit,robust) eqlist; else;
		lsq(maxit=3Dmit) eqlist;
	if @ifconv^=3D1; then; goto 999;
	analyz(noprint) addlist thlist;

	rescalc namD tlistD ielasD iseelasD ipredD isepredD istoreD=20
		idtestD idsysD 2 0 sigmaD invsigD coef ses coefa sesa;

	if wstoreD=3D0; then; goto 783;
	out '\livsmed\data\tsp\foodbnk2';
	dot dlist;
genr cwt. nw1.usDm;                   ?***;
	enddot;
	out;
783 if istoreD=3D0; then; goto 777;
out '\livsmed\data\aids\3eqn\ml\m\D-a';           ?***;
	store1 ielasD iseelasD ipredD isepredD idtestD idsysD tlistD 2 0=20
		coef ses coefa sesa;
	out;

777 if homsymD^=3D0; then; homsym namD ielasD iseelasD ipredD isepredD=20
		isthsD istoreD idtestD idsysD 2;


? LIKELIHOOD RATIO AND OTHER TESTS

	if (metT=3D0)|(iconvT^=3D1)|(irtest=3D0); then; goto 999;
	test namD pD loglD sigmaD invsigD pT loglT sigmaT invsigT 1;
	if istoreD=3D0; then; goto 999;
	out '\livsmed\data\aids\3eqn\ml\m\D-a';         ?***;
	dot lrtlist;
		keep .D;
	enddot;
	out;


? PROCEDURE HOMSYM

	proc homsym nlist ielas iseelas ipred isepred isths istore idtest=20
		idsys imet;

? HOMOGENEITY

	dot dlist2;
frml lh., g.1 =3D -(g.2 + g.3);               ?***;
	enddot;
	dot eqlist addlist lhomlist;
		copy . sv.;
	enddot;
	dot dlist;
		copy cwt. svcwt.;
		copy xx. svxx.;
	enddot;
	dot dlist2;
		eqsub eq. lhomlist;
	enddot;
	dot addlist;
		eqsub . lhomlist;
	enddot;
	list alist alist lhomlist;
	dot dlist;
		eqsub cwt. lhomlist;
		eqsub xx. lhomlist;
	enddot;

	if irobust =3D 1; then; lsq(maxit=3Dmit,wname=3D@covu,robust) eqlist;
		else; lsq(maxit=3Dmit,wname=3D@covu) eqlist;
	set iconvH =3D @ifconv; if iconvH^=3D1; then; goto 895;
	if imet=3D2; then; goto 771;
	analyz(noprint) addlist lhomlist;
	list tlistH tlistH1;
	goto 772;
771 analyz(noprint) addlist lhomlist thlist;
	list tlistH tlistH1 tlistH2;
772 set idt=3Didtest*idtesths; set ids =3D idsys*idsyshs;
	set iel=3Dielas*ielashs; set ixse=3Diseelas*iselashs;
	set ipr=3Dipred*ipredhs; set ipse=3Disepred*ispredhs;
	rescalc namH tlistH iel ixse ipr ipse istore idt ids imet 1=20
		sigmaH invsigH coef ses coefa sesa;

	if isths=3D0; then; goto 863;
	if imet=3D1; then;
out '\livsmed\data\aids\3eqn\ml\m\T-h';           ?***;
	if imet=3D2; then;
out '\livsmed\data\aids\3eqn\ml\m\D-h';           ?***;
	store1 iel ixse ipr ipse idt ids tlistH imet 1 coef ses coefa=20
		sesa;
	out;

? SYMMETRY (AND HOMOGENEITY)

863 dot dlist3;                   ?//*;
frml ls.2, g.2 =3D g2.;                 ?//*;
enddot;                       ?//*;
	dot dlist2;
		eqsub eq. lsymlist;
	enddot;
	dot addlist;
		eqsub . lsymlist;
	enddot;
	dot lhomlist;
		eqsub . lsymlist;
	enddot;
	list alist alist lsymlist;
	dot dlist;
		eqsub cwt. lsymlist;
		eqsub xx. lsymlist;
	enddot;

	if irobust =3D 1; then; lsq(maxit=3Dmit,wname=3DsigmaH,robust) eqlist;
	else; lsq(maxit=3Dmit,wname=3DsigmaH) eqlist;
	set iconvS =3D @ifconv; if iconvS^=3D1; then; goto 867;
	list tlistS tlistS2;
	if imet^=3D2; then; analyz(noprint) addlist lhomlist lsymlist;
	else; analyz(noprint) addlist lhomlist lsymlist thlist;

	rescalc namS tlistS iel ixse ipr ipse istore idt ids imet 2=20
		sigmaS invsigS coef ses coefa sesa;

	if isths=3D0; then; goto 867;
	if imet=3D1; then;
out '\livsmed\data\aids\3eqn\ml\m\T-hs';          ?***;
	if imet=3D2; then;
out '\livsmed\data\aids\3eqn\ml\m\D-hs';          ?***;
	store1 iel ixse ipr ipse idt ids tlistS imet 2 coef ses coefa=20
		sesa;
	out;

? LIKELIHOOD RATIO AND OTHER TESTS

867 if irtest=3D0; then; goto 893;
	dot nlist;
		test namH p. logl. sigma. invsig. pH loglH sigmaH invsigH 0;
		if iconvS=3D1; then;
		test namS p. logl. sigma. invsig. pS loglS sigmaS invsigS 1;
		if istore=3D0; then; goto 897;
		if imet=3D1; then;
out '\livsmed\data\aids\3eqn\ml\m\T-a';           ?***;
		if imet=3D2; then;
out '\livsmed\data\aids\3eqn\ml\m\D-a';           ?***;
		dot lrtlist;
			dot H S;
				if iconv.=3D1; then; keep ..;
			enddot;
		enddot;
		out;
897   set iii =3D 0;
	enddot;
	if iconvS^=3D1; then; goto 893;
	test namScH pH loglH sigmaH invsigH pS loglS sigmaS invsigS 1;
	if isths=3D0; then; goto 893;
	if imet=3D1; then;
out '\livsmed\data\aids\3eqn\ml\m\T-h';           ?***;
	if imet=3D2; then;
out '\livsmed\data\aids\3eqn\ml\m\D-h';           ?***;
	dot lrtlist;
		keep .ScH;
	enddot;
	out;

893 dot eqlist addlist lhomlist;
		copy sv. .;
	enddot;
	dot dlist;
		copy svcwt. cwt.;
		copy svxx. xx.;
	enddot;

895 list alist addlist;

	endproc;


? PROCEDURE TEST

	proc test name pU lU sigU invsigU pR lR sigR invsigR ik;

	dot name;
		set df. =3D pu - pr;
		set wald =3D wld.;
		set pwald =3D pwd.;
		set cwald =3D cwd.;
		set pcwald =3D pcw.;
		set df =3D df.;
		print df.;
	enddot;
	set dfw =3D ntot-pU;
	set dflm =3D ntot-pR;
	mat sigw =3D invsigu*sigr;
	mat siglm =3D invsigr*sigu;
	mat trW =3D tr(sigw);
	mat trLM =3D tr(siglm);
	set lrt =3D 2*(lu-lr);
	set ssrw=3Dnobs*(trW - n1);
	set ssrlm=3Dnobs*(n1 - trLM);

	set k1 =3D 1 - pu/ntot;
	set klm =3D 1 - pr/ntot;
	set clrt =3D k1*lrt;
	set cssrw =3D k1*ssrw;
	set cssrlm =3D klm*ssrlm;

	dot lrt clrt wald cwald ssrw cssrw ssrlm cssrlm;
		cdf(chisq,df=3Ddf) . p.;
		print . p. df;
	enddot;
	elrt lrt exlrt pexlrt df dfex2 n1 pU;
	print df dfex2 exlrt pexlrt;
	if ik=3D0; then; print exwldH pxwldH;

	print k1 klm;
	dot name;
		set lrt. =3D lrt; set clrt. =3D clrt; set exlrt.=3D exlrt;
		set plrt.=3Dplrt; set pclrt.=3Dpclrt; set pxlrt.=3Dpexlrt;
	enddot;
	endproc;


? PROCEDURE ELRT - "EXACT" LR TEST

	proc elrt llrt lex plex dfr df2 nn1 ppu;

	set lam =3D exp(-llrt/@nob);
	if (nn1^=3Ddfr)&(nn1^=3D1); then; goto 834;
	set sex =3D 1; goto 836;
834 if (nn1^=3D2)&(dfr^=3D2*nn1); then; goto 837;
	set sex=3D2; goto 836;
837 if dfr<=3D2; then; goto 835;
	set sex =3Dnn1*sqrt(((dfr**2)-4)/((dfr**2)+(nn1**4)-(5*(nn1**2))));
836 set gex =3D @nob - (ppu/nn1) - (nn1 - (dfr/nn1) + 1)/2;
	set rex =3D (dfr/2) - 1;
	set df2 =3D gex*sex - rex;
	set lex =3D (df2/dfr)*(lam**(-1/sex) - 1);

	set dl =3D int(df2);
	set du =3D ceil(df2);
	cdf(f,df1=3Ddfr,df2=3Ddl) lex plex;
	if dl=3Ddu; then; goto 628;
	cdf(f,df1=3Ddfr,df2=3Ddu) lex plrt2;
	set plex=3D(df2-dl)*plrt2 + (du-df2)*plex;
	goto 628;
835 set lex=3D@miss; set plex=3D@miss; set df2=3D@miss;
628 set iii=3D0;

	endproc;


? PROCEDURE RESCALC

	proc rescalc namlt tlist ielas iseelas ipred isepred istore=20
		idtest idsys imet inhs sigma invsig coef ses coefa sesa;

	dot coef ses coefa sesa;
		copy @. .;
	enddot;
	supres @logl @nob @coef @ses;
	set akaike =3D 2*(@ncid - @logl)/nobs;
	print akaike;
	set dfc =3D (@ncid/n1);
	set nobsc =3D nobs - dfc;
	set irr=3D0;
	dot dlist2;
		set irr =3D irr + 1;
		set rsq. =3D @rsq(irr);
		set dw. =3D @dw(irr);
	enddot;

	genr cwt1 what; uhat =3D w1 - what; uhat2 =3D uhat**2;
	msd(corr,noprint) w1 what;
	set rsq1 =3D (@corr(1,2))**2;
	msd(noprint) uhat2; set sume2 =3D @sum(1);
	smpl t2,tt; dwv =3D (uhat - uhat(-1))**2;
	msd(noprint) dwv;
	set dw1 =3D @sum(1)/sume2; smpl smplest;
	dot dlist;
		set arsq. =3D ((nobs-1)*rsq. + 1 - dfc)/nobsc;
		print rsq. arsq. dw.;
	enddot;

	dot namlt;
		set iii=3D0;
		set p. =3D @ncid; set logl =3D @logl; set logl. =3D logl;
		copy @covu sigma;
		mat invsig =3D sigma";
		mat dete =3D det(sigma);
		mat sigew =3D invsig*sigw;
		mat trew =3D tr(sigew);
		set rsqdet =3D 1 - (nobs/(nobs-1))*((dete/detw)**(1/n1));
		set arsqdet =3D ((nobs-1)*rsqdet + 1 - dfc)/nobsc;
		print rsqdet arsqdet;
		set rsqtrw =3D 1 - (nobs/(nobs-1))*(n1/trew);
		set arsqtrw =3D ((nobs-1)*rsqtrw + 1 - dfc)/nobsc;
		print rsqtrw arsqtrw;
		if (metT=3D0)|(inhs^=3D0)|(imet=3D1)|(iconvT^=3D1); then; goto 114;
		dot dlist;
			set rsq.cT. =3D (rsq. - rsqT.)/(1 - rsqT.);
			set ar2.cT. =3D (arsq. - arsqT.)/(1 - arsqT.);
			print rsq.cT. ar2.cT.;
		enddot;
		set r2det.cT =3D (rsqdet - rsqdetT)/(1 - rsqdetT);
		set ar2dt.cT =3D (arsqdet - arsqdetT)/(1 - arsqdetT);
		print r2det.cT ar2dt.cT;
		set r2trw.cT =3D (rsqtrw - rsqtrwT)/(1 - rsqtrwT);
		set ar2tr.cT =3D (arsqtrw - arsqtrwT)/(1 - arsqtrwT);
		print r2trw.cT ar2tr.cT;
114   if (irtest=3D0)|((inhs=3D2)&(imet=3D1)); then; goto 918;
		set dfex2 =3D nobsc - n + 2;
		set ikk=3D1;
		dot tlist;
			anz nam. .lt;
			if (inhs^=3D0)|(ikk^=3D1); then; goto 57;
			set exwld. =3D dfex2*wld./(nobs*dfex1);
			cdf(F, df1=3Ddfex1, df2=3Ddfex2) exwld. pexwald;
			print exwld. pexwald dfex1 dfex2;
			set pxwld. =3D pexwald;
57      set ikk =3D ikk + 1;
		enddot;
918   list lsqnm @rnms; rename @vcov lsqvcov;
		if idtest^=3D0; then; diagnost imet p. idsys;
		if ielas^=3D0; then; elas iseelas imet 0 lsqnm lsqvcov istore;
		if iseelas=3D0; then; set isse=3D0; else; set isse=3Disepred;
		if (ielas^=3D0)&(tdt>tet);then;
		elas isse imet 1 lsqnm lsqvcov istore;
		if ipred^=3D0; then; pred isepred istore 0 lsqnm lsqvcov;
		if (ipred^=3D0)&(tdt>tet); then;
		pred isepred istore 1 lsqnm lsqvcov;
	enddot;
	nosupres @coef @ses @logl @nob;
	endproc;


? PROCEDURE ANZ

	proc anz name eqlt;
	set dfcor =3D nobsc/nobs;
	analyz(noprint) eqlt;
	dot name;
	set wld. =3D @wald;
	set pwald =3D %wald; set pwd. =3D pwald;
	set cwald =3D wld.*dfcor; set cwd. =3D cwald;
	rename @ncida df;
	cdf(chisq,df=3Ddf) cwald pcwald;
	print wld. pwald cwald pcwald df;
	set pcw. =3D pcwald;
	enddot;
	endproc;


? PROCEDURE DIAGNOST

	proc diagnost imet np idsys;

	set rdfc =3D round(dfc);
	genr cpp phat;
	lxh =3D lx - phat;
	linch =3D linc - phat;
	lexph =3D lexp - phat;

	set eqnumb=3D0;
	if imet =3D 1; then; list vlist vlists;
	if imet =3D 2; then; list vlist vlistd;
	dot exph inch;
		olsq(silent) lxh vlist l.;
		rename @fit lx.;
	enddot;
	list vlist vlist lxh;

	dot dlist;
		set eqnumb =3D eqnumb + 1;
		print eqnumb;

		genr cwt. what;
		what2 =3D what**2;
		what3 =3D what**3;
		uhat =3D w. - what;
		uhat2 =3D uhat**2;
		uhat3 =3D uhat**3;
		uhat4 =3D uhat**4;
		wht. =3D what; wht2. =3D what2; wht3. =3D what3;
		uht. =3D uhat; uht2. =3D uhat2;

		frest 0 uhat vlist BG1lt namBG1. 1;
		frest 0 uhat vlist BG2lt namBG2. 2;
		frest 1 uhat2 clist BP1lt namBP1. 0;
		frest 1 uhat2 clist BP2lt namBP2. 0;
		frest 1 uhat2 clist ARCH1lt namAR1. 1;
		frest 1 uhat2 clist ARCH2lt namAR2. 2;
		frest 0 w. vlist RSET2lt namRS2. 0;
		frest 0 w. vlist RSET3lt namRS3. 0;
		frest 0 w. vlist HWUTClt namHTC. 0;
		frest 0 w. vlist HWUDIlt namHDI. 0;

		msd(noprint) uhat;
		dot 2 3 4;
			msd(noprint) uhat.;
			rename @mean mu.;
		enddot;
		set sk =3D (mu3**2)/(mu2**3);
		set kt =3D mu4/(mu2**2);
		set bj. =3D nobsc*(sk/6+((kt - 3)**2)/24);
		set sk2 =3D sk*(((nobsc-1)**2)/(nobsc-2));
		set kt2 =3D ((((nobsc-1)**3)*(nobsc+1))/(nobsc*(nobsc-2)*
				(nobsc-3)))*((kt-3*((nobsc-1)/(nobsc+1)))**2);
		set bjas. =3D sk2/6 + kt2/24;
		cdf(chisq,df=3D2) bjas. pbjas.;
		print bj. bjas. pbjas.;

		set uhatss =3D mu2*nobs;
		uhat2s=3D0;
		smpl t1; set tk=3Dt1+round(np/n1);
		uhat2s =3D uhat2;
		smpl t2,tt; uhat2s =3D uhat2s(-1) + uhat2;
		smpl tk,tt; trend trd;
		csq =3D abs((uhat2s/uhatss)-(trd/@nob));
		set mcsmsq =3D (@nob/2) - 1;
		msd(noprint) csq;
		set cusmsq. =3D @max;
		print cusmsq. mcsmsq;
		smpl smplest;
	enddot;

	if idsys=3D0; then; goto 781;
	dot parlst;
		copy . .s;
	enddot;
	dot equ2 eqw;
		lsq(maxit=3Dmit,silent) .lt;
		set ifc.=3D@ifconv;
		set l. =3D @logl;
		set p. =3D @ncid;
	enddot;
	smpl t2,tt;
	dot equ equ2;
		lsq(maxit=3Dmit,silent) .lt;
		set ifc1.=3D@ifconv;
		set l1. =3D @logl;
		set p1. =3D @ncid;
	enddot;
	smpl t3,tt;
	dot equ equ2;
		lsq(maxit=3Dmit,silent) .lt;
		set ifc2.=3D@ifconv;
		set l2. =3D @logl;
		set p2. =3D @ncid;
	enddot;
	smpl smplest;
	if ifc1equ=3D1; then; systst lstl11 l1equ p1equ 1;
	if ifc2equ=3D1; then; systst lstl12 l2equ p2equ 2;
	if ifcequ2=3D1; then; systst lstl30 lequ2 pequ2 0;
	if ifc1equ2=3D1; then; systst lstl31 l1equ2 p1equ2 1;
	if ifc2equ2=3D1; then; systst lstl32 l2equ2 p2equ2 2;
	if ifceqw=3D1; then; systst lstlst2 leqw peqw 0;
	dot parlst;
		copy .s .;
	enddot;
781 set iii=3D0;

	endproc;


? PROCEDURE FREST

	proc frest fnr y listv listx namelist ix;
	set tx =3D t1 + ix; smpl tx,tt;
	dot namelist;
		if fnr=3D1; then; regopt(pvcalc) fst;
		length listx npx; if npx=3D1; then; regopt(pvcalc) t;
		olsq(silent) y listx listv;
		set df2 =3D @nob - @ncid;
		if fnr=3D1; then; goto 997;
		if (npx=3D1)&(@ncid=3D@ncoef); then; goto 996;
		set ssru =3D @ssr; set npu =3D @ncid;
		olsq(silent) y listv;
		set dfr =3D npu - @ncid;
		if dfr=3D0; then; goto 998;
		set f. =3D ((@ssr - ssru)/dfr)/(ssru/df2);
		cdf(f,df1=3Ddfr,df2=3Ddf2) f. p.;
		goto 995;
997   set f. =3D @fst; set p.=3D%fst; set dfr=3D@ncid-1;
		regopt(nopvcalc) fst;
		goto 995;
998   set f.=3D@miss;set p.=3D@miss; goto 995;
996   set f. =3D @t(1)**2; set p.=3D%t(1); set dfr=3D1;
		regopt(nopvcalc) t;
995   print f. p. dfr df2;
	enddot;
	smpl smplest;

	endproc;


? PROCEDURE SYSTST

	proc systst listd lglr pr ix;
	set tx =3D t1 + ix; smpl tx,tt;
	dot listd;
		lsq(maxit=3Dmit,silent) .lt;
		if @ifconv^=3D1; then; goto 475;
		set lrt =3D 2*(@logl - lglr);
		set df. =3D @ncid - pr; set df =3D df.;
		print df.;
		if df=3D0; then; goto 475;
		cdf(chisq,df=3Ddf) lrt plrt;
		length .lt nn1;
		elrt lrt exlrt pexlrt df. df2 nn1 @ncid;
		set lr. =3D lrt; set pl. =3D plrt;
		set ex. =3D exlrt; set px. =3D pexlrt;
		goto 476;
475   dot lrt plrt df2 exlrt pexlrt;
			set . =3D @miss;
		enddot;
476   print lrt plrt df2 exlrt pexlrt;
		set iii=3D0;
	enddot;
	smpl smplest;
	endproc;


? PROCEDURE ELAS

	proc elas iseelas imet ipp lsqnm lsqvcov istore;

	if ipp=3D0; then; goto 712;
	copy smplpred sample; copy smplpred sample2;
	set ta1 =3D tt1; set tat =3D tdt; set ta0 =3D ta1;
	set te =3D tt1; set ti =3D 1;
	goto 713;;

712 copy smplest sample; copy smplest2 sample2;
	set ta1 =3D t1; set tat =3D tt; set ta0 =3D t0;
	set te =3D t0; set ti =3D 1;
	if iseelas.ge.2; then; set te=3Dtt;
	if iseelas=3D2; then; set ti =3D tt-t0;;

713 if imet^=3D2; then; goto 557;
mmake vth th11-th13 th21-th23 th31-th33;          ?***;
	mform(nrow=3Dn,ncol=3Dn) mth=3Dvth;
	matran mth mth;
	if ipp=3D0; then; print mth;
	mat invmith =3D (identn - mth)";

557 dot dlist;
		eqsub ce. cwt. alist;
eqsub cep.1 cwt. xx1 alist;               ?//*;
eqsub cep.2 cwt. xx2 alist;               ?//*;
eqsub cep.3 cwt. xx3 alist;               ?//*;
eqsub cpc.1 cwt1 cep.1 ce.;               ?//*;
eqsub cpc.2 cwt2 cep.2 ce.;               ?//*;
eqsub cpc.3 cwt3 cep.3 ce.;               ?//*;
	enddot;


? CALCULATE ELASTICITIES

	smpl sample2;
	dot dlist;
		dot dlist;
			genr cpc.. epc..;
			genr cep.. ep..;
		enddot;
		genr ce. e.;
	enddot;

	dot dlist;
	enddot;
	dot dgnl;
		ep. =3D ep. -1;
		epc. =3D epc. -1;
	enddot;


? CALCULATE STANDARD ERRORS

	do t =3D ta0 to te by ti;
		smpl t t;
		set iii=3D0;
		dot dlist;
			dot dlist;
				if iseelas=3D0; then; goto 111;
				analyz(noprint,names=3Dlsqnm,vcov=3Dlsqvcov) cep..;
				set vpt.. =3D @sesa(1);
				analyz(noprint,names=3Dlsqnm,vcov=3Dlsqvcov) cpc..;
				set vpct.. =3D @sesa(1);
				set sp..(t)=3Dvpt..; set spc..(t)=3Dvpct..;
111       set ept..=3Dep..(t); set epct..=3Depc..(t);
			enddot;
			if iseelas=3D0; then; goto 112;
			analyz(noprint,names=3Dlsqnm,vcov=3Dlsqvcov) ce.;
			set vt. =3D @sesa(1);
			set se.(t)=3Dvt.;
112     set et.=3De.(t);
		enddot;

mmake ve et1-et3;                   ?***;
mmake vep ept11-ept13 ept21-ept23 ept31-ept33;        ?***;
mmake vepc epct11-epct13 epct21-epct23 epct31-epct33;     ?***;
		dot ep epc ;
			mform(nrow=3Dn,ncol=3Dn) m.=3Dv.;
			matran m. m.;
		enddot;
		mform(nrow=3Dn,ncol=3D1) me=3Dve;

		if imet^=3D2; then; go to 131;
		dot e ep epc;
			mat long. =3D invmith*m.;
		enddot;

131   if (t=3Dta0)&(ipp=3D0); then; goto 143;
		if (iseelas=3D2)&(ipp=3D0); then; goto 149;
		if iprint=3D0; then; goto 147;
149   if (iprint=3D2)|(t=3Dte)|(ipp=3D1); then; print t;
		goto 146;
143   print atmean;

146   matran ve ve;
		if (iprint^=3D2)&(t^=3Dta0)&(t^=3Dte)&(ipp=3D0); then; goto 144;
		dot ve mep mepc;
			print .;
		enddot;

144   if iseelas=3D0; then; goto 148;
mmake vv vt1-vt3;                   ?***;
mmake vvp vpt11-vpt13 vpt21-vpt23 vpt31-vpt33;        ?***;
mmake vvpc vpct11-vpct13 vpct21-vpct23 vpct31-vpct33;     ?***;
		dot vp vpc ;
			mform(nrow=3Dn,ncol=3Dn) m.=3Dv.;
			matran m. m.;
		enddot;

		matran vv vv;
		if (iprint^=3D2)&(t^=3Dta0)&(t^=3Dte)&(ipp=3D0); then; goto 148;
		dot vv mvp mvpc;
			print .;
		enddot;

148   if imet=3D1; then; goto 147;
		matran longe longe;
		if (iprint^=3D2)&(t^=3Dta0)&(t^=3Dte)&(ipp=3D0); then; goto 147;
		dot longe longep longepc;
			print .;
		enddot;

147   if(ipp^=3D0)|(istore=3D0); then; goto 344;
		if iseelas=3D0; then; list t0list ve mep mepc;;
		else; list t0list ve mep mepc vv mvp mvpc;
		if iseelas^=3D1; then; list t1list t0list;
		else; list t1list ve mep mepc;
		if t^=3Dta0; then; goto 348;
		dot t0list;
			copy . b.;
		enddot;
348   if t^=3Dte; then; goto 344;
		dot t1list;
			copy . l.;
		enddot;
344   set iii=3D0;
	enddo;


? PRINT ELASTICITIES

	smpl sample;
	if (iprint=3D0)&(@nob^=3D1); then; goto 118;
	if ((ipp=3D0)&(iseelas^=3D3))|((ipp=3D1)&(iseelas=3D0)); then; goto =
113;
	dot dlist;
		cilo =3D e. - critval * se.;
		cihi =3D e. + critval * se.;
		print e. se. cilo cihi;
		dot dlist;
			cilo =3D ep.. - critval * sp..;
			cihi =3D ep.. + critval * sp..;
			print ep.. sp.. cilo cihi;
			cilo =3D epc.. - critval * spc..;
			cihi =3D epc.. + critval * spc..;
			print epc.. spc.. cilo cihi;
		enddot;
	enddot;
	goto 118;

113 dot dlist;
		print e. ep.lst;
		print epc.lst;
	enddot;

118 if @nob=3D1; then; goto 117;
	dot dlist;
		msd e. ep.lst epc.lst;
		if ((ipp=3D0)&(iseelas=3D3))|((ipp=3D1)&(iseelas^=3D0)); then;
		msd se. sp.lst spc.lst;
	enddot;

117 smpl smplest;
	endproc;


? PROCEDURE PRED

	proc pred isepred istore ipp lsqnm lsqvcov;

	if ((ipp=3D0)&(iprint=3D0)&(istore=3D0)); then; goto 336;
	if ipp=3D0; then; goto 333;
	if isepred=3D0; then; set ise=3D0; else; set ise=3D1;
	copy smplpred sample; set tf=3Dtt1; set ts=3Dtdt; print predper;
	goto 334;
333 if isepred^=3D2; then; set ise=3D0; else; set ise=3D1;
	copy smplest sample; set tf=3Dt1; set ts=3Dtt;
	if iprint^=3D0; then; print samplper;

334 smpl sample;
	dot dlist;
		genr cwt. wt.; cpt. =3D wt.*x; ct. =3D cpt.*npop;
	enddot;
	if ipp=3D0; then; goto 338;
	dot w cp c;
		dot dlist;
			.tdl. =3D (.. - .t.)**2;
			msd(noprint) .tdl.;
			set .tmsq. =3D @mean(1);
			set .trms. =3D sqrt(.tmsq.);
			set pr2.t. =3D 1 - (.tmsq./v..);
		enddot;
	enddot;

338 if ise=3D0; then; goto 335;
	do t=3Dtf,ts;
		smpl t,t;
		dot dlist;
			analyz(noprint,names=3Dlsqnm,vcov=3Dlsqvcov) cwt.;
			set swt.(t) =3D @sesa(1);
			set scpt.(t) =3D x * swt.(t);
			set sct.(t) =3D npop * scpt.(t);
		enddot;
	enddo;
	smpl sample;
	if (ipp=3D0)&(iprint=3D0); then; goto 336;
	dot w cp c;
		dot dlist;
			cilo =3D .t. - critval * s.t.;
			cihi =3D .t. + critval * s.t.;
			print .t. s.t. cilo cihi ..;
			if ipp^=3D0; then; print .trms. pr2.t.;
		enddot;
	enddot;
	goto 336;

335 if (ipp=3D0)&(iprint=3D0); then; goto 336;
	dot w cp c;
		dot dlist;
			print .t. ..;
			if ipp^=3D0; then; print .trms. pr2.t.;
		enddot;
	enddot;

336 smpl smplest;
	endproc;


? PROCEDURE STORE1

	proc store1 ielas iseelas ipred isepred idtest idsys testlist=20
		imet inhs cvec svec cavec csvec;

	keep smplest smplest2 smplall smplall2;
	keep parlst cvec svec cavec csvec;
	if (metT=3D0)|(inhs=3D^0)|(iconvT^1)|(imet=3D1); then; goto 274;
	keep r2detDcT ar2dtDcT r2trwDcT ar2trDcT;
	dot dlist;
		keep rsqDcT. ar2DcT.;
	enddot;

274 if idtest=3D0; then; goto 275;
	keep mcsmsq;
	dot dtlist;
		dot dlist;
			keep f.. p..;
		enddot;
	enddot;
	dot BJ BJAS PBJAS CUSMSQ;
		dot dlist;
			keep ..;
		enddot;
	enddot;

	if idsys=3D0; then; goto 275;
	dot syslt;
		keep ex. px. lr. pl.;
	enddot;

275 if irtest=3D0; then; goto 273;
	dot wld pwd cwd pcw;
		dot testlist;
			keep ..;
		enddot;
	enddot;
	if inhs=3D0; then; keep exwldH pxwldH;

273 smpl smplest;
	if ielas=3D0; then; goto 276;
	keep elaslt;
	if iseelas=3D3; then; keep selaslt;
	if iseelas=3D0; then; list t0list ve mep mepc;;
	else; list t0list ve mep mepc vv mvp mvpc;
	if iseelas^=3D1; then; list t1list t0list;
	else; list t1list ve mep mepc;
	dot t0list;
		keep b.;
	enddot;
	dot t1list;
		keep l.;
	enddot;
276 if ipred=3D0; then; goto 277;
	smpl smplall;
	keep predlst varlst;
	if isepred=3D2; then; keep spredlst;
	dot w cp c;
		dot dlist;
			keep pr2.t.; keep .trms.;
		enddot;
	enddot;
277 if tdt<=3Dtet; then; goto 278;
	keep smplpred;
	smpl smplpred;
	if ielast=3D0; then; goto 279;
	keep elaslt;
	if isepred^=3D0; then; keep selaslt;
	goto 278;
279 if (ipred^=3D0)&(isepred=3D1); then; keep spredlst;
278 smpl smplest;

	endproc;


999 END;

------_=_NextPart_000_01BE4DD7.91F95EA0--

---------- End of message ----------

From: "Krassimir Petrov" <krassimir_petrov@hotmail.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: RATS for estimation of AIDS using cross-section data
Date: Mon, 01 Feb 1999 04:18:22 PST
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Mime-Version: 1.0
Content-Type: text/plain
X-Mailer: Mercury MTS (Bindery) v1.40




>From: Karin Elsner <elsner@iamo.uni-halle.de>
>To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
>Subject: RATS for estimation of AIDS using cross-section data
>Date: Mon, 01 Feb 1999 09:39:45 +0100
>Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
>
>Dear colleagues,
>I am using RATS for doing cross-sectional analysis because this 
software
>is available in the institute I am working at. More specifically, I
>estimate an Almost Ideal Demand System including socio-demographic
>variables.
>I face a lot of problems in transferring the algebraic model and the
>estimation procedures necessary for cross-section analysis (e.g.
>modified Heckamn procedure to account for zero expenditures, and so on)
>into the RATS programme syntax. In order to avoid inventing the wheel a
>hundredth time, I would like to ask whether somebody has estimated an
>AIDS with RATS using cross-sectional or also time-series data and
>whether he or she could imagine giving me the respective programme 
file.
>I know that this file would in no case be self-contained but that there
>will be notable differences with respect to my model specification and
>special dataset. But maybe the code could serve as a template for my 
own
>analysis and give me some ideas on how to solve the problems I am
>facing.
>Thank you for your assistance,
>Karin.
>-- 
>*********************************************************************
>                                                                      
>  Karin Elsner             Institute of Agricultural Development      
>                           in Central and Eastern Europe (IAMO)       
>                           Emil-Abderhalden-Strasse 20                
>                           06108 Halle                                
>                           Germany                                    
>                                                                      
>                           --------------------------------------     
>                                                                      
>                           Phone +49 345 / 5 52 29 33                 
>                           Fax   +49 345 / 5 00 81 77                 
>                           Email elsner@iamo.uni-halle.de                
>                    http://www.landw.uni-halle.de/iamo/iamo_e.htm     
>                                                                      
>*********************************************************************
>


Dear Karen,

when I took a PhD level course in Agricultural Demand Economics a few 
years ago everyone in my class had to estimate an AIDS model as one out 
of 5 homework assignments. The majority of my classmates did it in SAS, 
some in RATS, few in GAUSS, and I - in MATLAB.

I don't keep my code from a couple of years ago, but I do remember well 
that this was a no-brain straightforward code writing with a minimum 
effort of time. Hey, it was just another homework!

So, my advice is: go ahead and reinvent that wheel !!

Chris.
PhD Candidate.
The Ohio State University

______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com

---------- End of message ----------

From: Norman Morin <m1njm00@frb.gov>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: generation data
Date: Mon, 01 Feb 1999 07:57:18 -0500
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-version: 1.0
X-Mailer: exmh version 2.0.2.4 7/8/98 (via Mercury MTS (Bindery) v1.40)
Content-type: text/plain; charset=us-ascii


> Does anyone kwno how to generate data (simulate) on RATS.
> I would like to ganerate 100 observation of the following model:
> 
>                y(t)  =  0.8*y(t-1) + e
> 
> being "e" a random error N ~ (0, 2).
> 

RATS' SIMULATE command is great for this sort of thing.  But for
a quick simulation using only one lag, the easiest way is (assuming
y(0) = 0.):

SET(first=0.) y = 0.8*y{1} + %ran(sqrt(2.))

or you can do

SET e = %ran(sqrt(2.))
SET(first=0.) y = 0.8*y{1} + e

Norm
-- 

                           Norman J. Morin 
                  nmorin@frb.gov or m1njm00@frb.gov         
          --------------------------------------------------
          Division of Research and Statistics * Mail Stop 82
           Board of Governors of the Federal Reserve System 
               Washington, D.C. 20551 * (202) 452-2476
          --------------------------------------------------



---------- End of message ----------

From: Ricardo Chaves Lima <rcl@netpe.com.br>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: generation data
Date: Mon, 01 Feb 1999 18:36:00 -0300
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
X-Mailer: Mozilla 4.01 [en] (Win95; I) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Thanks Norman!

Ricardo


---------- End of message ----------

From: Karin Elsner <elsner@iamo.uni-halle.de>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: RATS for estimation of AIDS & Generalized Heckman Procedure
Date: Tue, 02 Feb 1999 09:28:00 +0100
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Organization: Institut für Agrarentwicklung
X-Mailer: Mozilla 3.01 [de] (Win95; I) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Dear collegues,
thanks to all of you who quickly gave me support, helped me to fix ideas
or concretely assisted me. I do not feel any longer as single fighter
despite nobody here could give me concrete advice. 

Special thanks to David Edgerton, I safely received your files. I
already know your book, and it has been and still is of great use for
me.

For those of you that are interested in the modified Heckman procedure,
here some articles. There you find as well hints for further reading. Or
search in JEL CD-ROM, using the keyword "Generalized Heckman
Procedure".

Heien, D., and Durham, C. (1991): A Test of the Habit Formation
Hypothesis Using Household Data. Review of Economics and Statistics,
Vol.73, May, pp.189-199.

Heien, D., and Roheim Wessels, C.(1990): Demand Systems Estimation with
Microdata: A Censored Regression Approach. Journal of Business and
Economics Statistics, Vol.8, No.3, pp.365-371.

Park, John L., Rodney B. Holcomb, Kellie Curry Raper und Oral Capps, Jr.
(1996): A Demand Systems Analysis of Food Commodities by U.S. Households
Segmented by Income. American Journal of Agricultural Economics, Vol.78,
pp.290-300. 

I hope that helps. Thank you again,
Karin.
-- 
*********************************************************************
                                                                      
  Karin Elsner             Institute of Agricultural Development      
                           in Central and Eastern Europe (IAMO)       
                           Emil-Abderhalden-Strasse 20                
                           06108 Halle                                
                           Germany                                    
                                                                      
                           --------------------------------------     
                                                                      
                           Phone +49 345 / 5 52 29 33                 
                           Fax   +49 345 / 5 00 81 77                 
                           Email elsner@iamo.uni-halle.de                
                    http://www.landw.uni-halle.de/iamo/iamo_e.htm     
                                                                      
*********************************************************************

---------- End of message ----------

From: mike mccracken <mmccrac@unix1.sncc.lsu.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: kernel-based prediction
Date: Tue, 02 Feb 1999 17:10:39 -0600
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
X-Mailer: QUALCOMM Windows Eudora Pro Version 3.0.3 (32) (via Mercury MTS (Bindery) v1.40)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"

Fellow fRAT members:
	I need to construct a sequence of kernel-based nonparametric forecasts of
the Nadarya-Watson type (not LWR-though that would be interesting as well).
 Has anyone used RATS to do this?  Surely something akin to the source code
"Kernel.src" (but for constructing regression functions) exists somewhere.
Thanks for the help.

	Mike

---------- End of message ----------

From: "Mustafa Ozer" <ozer@vm.baum.anadolu.edu.tr>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: 
Date: Fri, 5 Feb 1999 10:50:48 +0200
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
Content-Type: text/plain;
Content-Transfer-Encoding: 7bit
X-Mailer: Microsoft Outlook Express 4.72.2106.4 (via Mercury MTS (Bindery) v1.40)

Dear Friends,
I would like to know if I can test structural break in some time series
employing Zivot and Andrew model using RATS. In addition, Can I perform
Gregory and Hansen cointegration test on RATS. Thanks for your valuable
remarks. Mustafa OZER


---------- End of message ----------

From: Ricardo Chaves Lima <rcl@netpe.com.br>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: help on Garch/Arch with Kalman filter
Date: Fri, 05 Feb 1999 21:12:43 -0300
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
X-Mailer: Mozilla 4.01 [en] (Win95; I) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Dear Fellows,

Does anyone know a RAT's program (or procedure) to run Garch/Arch
model using Kalman Filter.

thanks!

Ricardo Chaves Lima
Depaartment of Economics
Universidade Federal de Pernambuco, Brazil


---------- End of message ----------

From: "chakib tahiri" <chakib@acdim.co.ma>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Recursive Program
Date: Mon, 8 Feb 1999 11:13:52 -0000
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
X-Mailer: Microsoft Internet Mail 4.70.1155 (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

Dear fellows,
I am trying to write a recursive program that computes the sequence of
numbers generated by non-linear difference equations. To give an example,
suppose the following equation:  A(n+1) = (A(n) + 2) / (A(n) + 1) given
initial condition, say A(0) =  10. 
I write the following program:
all 0 20
dec real count A
compute count = 0
compute A = 10
while count < 20 {
compute Aa = (A + 2) / (Aa + 1)
compute count = count + 1
display 'A = ' A
}
end while

This program works as it gives me the correct sequence of numbers up to the
allocate parameter whis is set at 20 terms, however, the sequence so
generated is not a series, so I cannot use it for graphics for example.
My question is how would I transform that program so that the sequence of
number generated is a series.
thank you very much

Chakib Tahiri
University of Marrakech
Dept of Economics
Morocco.

---------- End of message ----------

From: Christopher F Baum <baum@bc.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Recursive Program
Date: Mon, 08 Feb 1999 08:49:10 -0500
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
X-Mailer: Mulberry (MacOS) [1.4.0, s/n P020-300786-009] (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Just use a vector rather than a scalar for Aa; 

dec vec real aa(20)
comp aa(1)=10
do i=2,20
comp aa(i)=(aa(i-1)+2.0)/(aa(i-1)+1.0)
dis i aa(i)
enddo

Does a pretty good job of locating sqrt(2.0).

Kit Baum
Boston College Economics

--On Mon, Feb 8, 1999 11:13 +0000 chakib tahiri <chakib@acdim.co.ma> wrote:

> Dear fellows,
> I am trying to write a recursive program that computes the sequence of
> numbers generated by non-linear difference equations. To give an example,
> suppose the following equation:  A(n+1) = (A(n) + 2) / (A(n) + 1) given
> initial condition, say A(0) =  10. 
> I write the following program:
> all 0 20
> dec real count A
> compute count = 0
> compute A = 10
> while count < 20 {
> compute Aa = (A + 2) / (Aa + 1)
> compute count = count + 1
> display 'A = ' A
> }
> end while
> 
> This program works as it gives me the correct sequence of numbers up to
> the allocate parameter whis is set at 20 terms, however, the sequence so
> generated is not a series, so I cannot use it for graphics for example.
> My question is how would I transform that program so that the sequence of
> number generated is a series.
> thank you very much
> 
> Chakib Tahiri
> University of Marrakech
> Dept of Economics
> Morocco.





---------- End of message ----------

From: Stein Gabriel <gabriel.stein@lombard-st.co.uk>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Baxter-King bandpass filter
Date: Wed, 10 Feb 1999 08:51:05 -0000
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2232.9) (via Mercury MTS (Bindery) v1.40)
Content-Type: text/plain

Has anyone written an instruction enabling RATS to use the Baxter-King
bandpass filter? Or anyone who knows how to do it?
Thanks
Gabriel
gabriel.stein@lombard-st.co.uk
 
Lombard Street Research's material is intended to encourage better
understanding of economic policy and financial markets. It does not
constitute a solicitation for the purchase or sale of any commodities,
securities or investments. Although the information compiled herein is
considered reliable, its accuracy is not guaranteed. Any person using
Lombard Street Research's material does so solely at his own risk and
Lombard Street Research shall be under no liability whatsoever in respect
thereof.


---------- End of message ----------

From: "Iacoviello,M  (pg)" <M.Iacoviello@lse.ac.uk>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: VAR and cointegration
Date: Wed, 10 Feb 1999 20:35:29 -0000
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2448.0) (via Mercury MTS (Bindery) v1.40)
Content-Type: text/plain

Dear RATS users,

does any of you have a routine to perform the kind of analysis done by King
Plosser Stock and Watson in their 1991 paper (stochastic trends and economic
fluctuations?)

The latest VAR routine is quite helpful since it allows me to do the
Blanchard Quah decomposition, but I cannot specify a non-triangular matrix.

Therefore I would be grateful if any of you could help in solving this
problem.


Yours faithfully

Matteo Iacoviello
Research Students
LSE
Houghton Street WC2A 2AE
London


---------- End of message ----------

From: "Holloway, James" <jholloway@treasury.gov.au>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: 2-axis graph
Date: Fri, 12 Feb 1999 20:07:25 +1100
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2232.9) (via Mercury MTS (Bindery) v1.40)
Content-Type: text/plain

Can anyone help me work out how to produce a graph with 2-axis?  

I have tried the following and the graph looks jumbled.

smpl 89:3 98:4
spgraph(twoscale) -10 to 10
graph(header="header", key=below)  2
# series1 
# series2 

graph 1
# series3
spgraph(done)



---------- End of message ----------

From: "Jimmy Lee" <jlee@tiger.fhsu.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: 2-axis graph
Date: Fri, 12 Feb 1999 09:31:30 -0600
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Mime-Version: 1.0
Content-type: text/plain; charset=us-ascii
X-Mailer: Mercury MTS (Bindery) v1.40



To fit two series in the same graph, the option GRAPH(OVERLAY=LINE)
generally works well.

Good Luck,
Jim Lee




"Holloway, James" <jholloway@treasury.gov.au> on 02/12/99 03:07:25 AM

Please respond to "RATS Discussion List" <RATS-L@efs.mq.edu.au>
                                                              
                                                              
                                                              
 To:      "RATS Discussion List" <RATS-L@efs.mq.edu.au>       
                                                              
 cc:      (bcc: Jimmy Lee/FHSU)                               
                                                              
                                                              
                                                              
 Subject: 2-axis graph                                        
                                                              








Can anyone help me work out how to produce a graph with 2-axis?

I have tried the following and the graph looks jumbled.

smpl 89:3 98:4
spgraph(twoscale) -10 to 10
graph(header="header", key=below)  2
# series1
# series2

graph 1
# series3
spgraph(done)






---------- End of message ----------

From: szuniga@entelchile.net (SERGIO ZUNIGA  JARA)
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Recursive Least Squares
Date: Tue, 16 Feb 1999 02:08:23 -0600
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
Content-Type: multipart/alternative;
X-Mailer: Microsoft Outlook Express 4.72.3110.5 (via Mercury MTS (Bindery) v1.40)

This is a multi-part message in MIME format.

------=_NextPart_000_0021_01BE5951.3B135BA0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Hello:
    Does anybody help me in order to perform recursive least squares in =
RATS? (Like Micro-TSP Plots for Recursive Residuals, CUSUM test, CUSUM =
of Squares Test, One-step forecast F test, N-step forecast F test and =
Recursive Coeff. estimates).
    I will appreciate if anyone give me a reference book that includes =
these matters.
    Regards,


***************************************************
Sergio Zuniga       szuniga@entelchile.net             =20
Universidad Catolica del Norte         =20
Coquimbo - Chile                           =20
Tel.: 56-51-327248                       =20
***************************************************

------=_NextPart_000_0021_01BE5951.3B135BA0
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD W3 HTML//EN">
<HTML>
<HEAD>

<META content=3Dtext/html;charset=3Diso-8859-1 =
http-equiv=3DContent-Type>
<META content=3D'"MSHTML 4.72.3110.7"' name=3DGENERATOR>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT color=3D#000000 size=3D2>Hello:</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2>&nbsp;&nbsp;&nbsp; Does anybody help =
me in order=20
to perform recursive least squares in RATS? (Like Micro-TSP Plots for =
Recursive=20
Residuals, CUSUM test, CUSUM of Squares Test, One-step forecast F test, =
N-step=20
forecast F test and Recursive Coeff. estimates).</FONT></DIV>
<DIV>&nbsp;&nbsp;&nbsp; I will appreciate if anyone give me a reference =
book=20
that includes these matters.</DIV>
<DIV><FONT color=3D#000000 size=3D2>&nbsp;&nbsp;&nbsp; <FONT=20
size=3D3>Regards,</FONT></FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT color=3D#000000=20
size=3D2>***************************************************<BR>Sergio=20
Zuniga&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; <A=20
href=3D"mailto:szuniga@entelchile.net">szuniga@entelchile.net</A>&nbsp;&n=
bsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=20
<BR>Universidad Catolica del=20
Norte&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; <BR>Coquimbo =
-=20
Chile&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=
bsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;&nbsp;=20
<BR>Tel.:=20
56-51-327248&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=
bsp;=20
<BR>***************************************************</FONT></DIV></BOD=
Y></HTML>

------=_NextPart_000_0021_01BE5951.3B135BA0--


---------- End of message ----------

From: Blanchard Pierre <blanchard@univ-paris12.fr>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Sure estimation
Date: Wed, 17 Feb 1999 18:11:43 +0100
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
X-Mailer: QUALCOMM Windows Eudora Pro Version 4.0.1  (via Mercury MTS (Bindery) v1.40)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"

Dear Rats users:

I'm looking for code for sure/maximum likelihood estimation  of translog
share equations (with symmetry and linear homogeneity in prices).
Thanks for your help.

Cheers,

Pierre Blanchard
Pierre Blanchard
Universite Paris XII Val de Marne
Faculte de Sciences Economiques et de Gestion
58 Av. Didier 
94214 Cedex La Varenne Saint-Hilaire
France
tel : (33) 01 49 76 80 50
fax: (33) 01 48 85 29 93
Email: blanchard@univ-paris12.fr

---------- End of message ----------

From: szuniga@entelchile.net (SERGIO ZUNIGA  JARA)
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Recursive Least Squares
Date: Wed, 17 Feb 1999 23:05:34 -0600
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
Content-Type: multipart/alternative;
X-Mailer: Microsoft Outlook Express 4.72.3110.5 (via Mercury MTS (Bindery) v1.40)

This is a multi-part message in MIME format.

------=_NextPart_000_0064_01BE5ACA.064A9F80
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Hello:
    Does anybody help me in order to perform recursive least squares in =
RATS? (Like Micro-TSP Plots for Recursive Residuals, CUSUM test, CUSUM =
of Squares Test, One-step forecast F test, N-step forecast F test and =
Recursive Coeff. estimates).
    I will appreciate if anyone give me a reference book that includes =
these matters.
    Regards,


***************************************************
Sergio Zuniga       szuniga@entelchile.net             =20
Universidad Catolica del Norte         =20
Coquimbo - Chile                           =20
Tel.: 56-51-327248                       =20
***************************************************

------=_NextPart_000_0064_01BE5ACA.064A9F80
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD W3 HTML//EN">
<HTML>
<HEAD>

<META content=3Dtext/html;charset=3Diso-8859-1 =
http-equiv=3DContent-Type><!DOCTYPE HTML PUBLIC "-//W3C//DTD W3 =
HTML//EN">
<META content=3D'"MSHTML 4.72.3110.7"' name=3DGENERATOR>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT color=3D#000000 size=3D2>Hello:</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2>&nbsp;&nbsp;&nbsp; Does anybody help =
me in order=20
to perform recursive least squares in RATS? (Like Micro-TSP Plots for =
Recursive=20
Residuals, CUSUM test, CUSUM of Squares Test, One-step forecast F test, =
N-step=20
forecast F test and Recursive Coeff. estimates).</FONT></DIV>
<DIV>&nbsp;&nbsp;&nbsp; I will appreciate if anyone give me a reference =
book=20
that includes these matters.</DIV>
<DIV><FONT color=3D#000000 size=3D2>&nbsp;&nbsp;&nbsp; <FONT=20
size=3D3>Regards,</FONT></FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT color=3D#000000=20
size=3D2>***************************************************<BR>Sergio=20
Zuniga&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; <A=20
href=3D"mailto:szuniga@entelchile.net">szuniga@entelchile.net</A>&nbsp;&n=
bsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=20
<BR>Universidad Catolica del=20
Norte&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; <BR>Coquimbo =
-=20
Chile&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=
bsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;&nbsp;=20
<BR>Tel.:=20
56-51-327248&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=
bsp;=20
<BR>***************************************************</FONT></DIV></BOD=
Y></HTML>

------=_NextPart_000_0064_01BE5ACA.064A9F80--


---------- End of message ----------

From: Irida Achthoven <achthovi@qed.econ.queensu.ca>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: garch standardized residuals
Date: Thu, 18 Feb 1999 07:02:03 -0800
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Organization: Queen's University
X-Mailer: Mozilla 3.04 (Win16; I) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Hi, 
I am a new subsriber and I have search the archives to find the answer
to my questions before posting here.
unfortunately i have noticed that these questions are asked frequently,
but there are no responses tothem. (at least not in the archives)

I hope that somebody will reponds , because I really need the help
1. I have astimeted the garch(p,g) model. I know that the residuals and
the  conditional covariance are resp sotred in the series U(T) and H(t)
but they are not accesible.
HOW CAN I CALCULATE THE STANDARDIZED ERRORS
HOW CAN I CALCULATE THE MOMENTS(KURTOSIS) OF THE STANDARD ERRORS.
HOW CAN I CALCULATE THE LJUNG BOX TEST STAT. FOR SERIAL CORRELATION.

i HAVE TRIED TO USE THE GARCH SRC CODE, SINCE IT IS A VERY VALUABLE CODE
BUT I HAVE RUN INTO A PROBLEM , AND DO NOT KNOW HOW TO SOLVE IT, THAT'S
WHY I TRIED THE CONVENTIONAL WAY

HERE IS THE APRT AND THE DISPLAYED ERROR CODE,  I GOT WHEN IMPLEMENTING
GARCH SRC

calendar(daily) 1994 1 4
all 0 1999:1:20
open data a:stock5.txt
data(format=free,org=obs) / apple gate hewl micro intel m fr hola hol
holp
SET Lapple = log(apple)
set Lgate =  log(gate)
set Lhewl =  log(hewl)
set Lmicro = log(micro)
set Lintel = log(intel)
set Dapple = Lapple -Lapple{1}
set Dhewl = Lhewl- Lhewl{1}
set Dgate  = Lgate- Lgate{1}
set Dmicro = Lmicro- Lmicro{1}
set Dintel = Lintel- Lintel{1}
set y = dmicro
source(noecho) c:\garch.src
@garch(mod=arch,dist=ged,ar=1,p=1,q=0,iter=100,cc=0.08, qc1=0.08,bhhh)
94:01;05  98:06:22  y nresids  cvar

## SX22. Expected Type SERIES, Got REAL Instead
>>>>ev=levc,l1=lc,m=mc<<<<

I HOPE SOMEBODY CAN TELL ME HOW TO GO ABOUT THE 3 FIRST QUESTIONS OR
TELL ME WHAT I DO WRONG IN THE ABOVE.

THANK YOU ,
IRIDA
END OF MESSAGE

---------- End of message ----------

From: Irida Achthoven <achthovi@qed.econ.queensu.ca>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: ARCH,( G) ARCH-M BOOTSTRAP FOR FORECASTING
Date: Thu, 18 Feb 1999 07:23:43 -0800
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Organization: Queen's University
X-Mailer: Mozilla 3.04 (Win16; I) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

HI EVERYBODY
I HOPE SOME BODY CAN HELP ME WITH ASAMPLE CODE FOR GENERATINGN FORECAST 
AND PREDICTION INTERVALS FOR ARCH- TYPE MODELS USING BOOTSTRAP
METHODOLOGY'

A MONTECARLO SIMULATION IS ALSO WELCOME.

THE SERIES I AM ESTIMATING IS STOCK RETURN SERIES.
MEAN EGN. Y(t)= B0 + Y(t-1) + E(T)
WHERE E(T) IS GARCH , FOR GARCH MODEL

THANK YOU

IRIDA
END OF MESSAGE

---------- End of message ----------

From: "Philippe PROTIN" <PROTIN@esa.upmf-grenoble.fr>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: garch standardized residuals
Date: Thu, 18 Feb 1999 15:40:26 +0200
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Organization: Ecole Superieure des Affaires
MIME-Version: 1.0
Content-type: text/plain; charset=ISO-8859-1
Content-transfer-encoding: Quoted-printable
X-mailer: Pegasus Mail for Windows (v2.54) (via Mercury MTS (Bindery) v1.40)

Dear Irida,

Here is a procedure I wrote to test residuals : standardized ans 
squared standardized residuals (LB STAT AND BERA JARQUE TEST FOR 
NOMALITY...)

Hope it will help.

        \\\|||||///
        \\  - -  //
         (  @ @  )
+------oOOo-(_)-oOOo---- +----------------------------------------+
|                        | Philippe PROTIN                             |
| Have a nice day !      | Ecole Sup=E9rieure des Affaires               |
|                        | Tel: (33) 04.76.82.57.48                    |
|                        | e-Mail: protin@esa.upmf-grenoble.fr         |  =
                              
+---------------Oooo-----+---------------------------------------------+
        oooO   (   )
       (   )    ) /
        \ (    (_/
         \_)

---------- End of message ----------

From: "Philippe PROTIN" <PROTIN@esa.upmf-grenoble.fr>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: garch standardized residuals
Date: Thu, 18 Feb 1999 15:40:26 +0200
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Organization: Ecole Superieure des Affaires
MIME-Version: 1.0
Content-type: Multipart/Mixed; boundary=Message-Boundary-18041
X-finfo: DOS,"testres.src",,,,Unknown
X-mailer: Pegasus Mail for Windows (v2.54) (via Mercury MTS (Bindery) v1.40)

--Message-Boundary-18041
Content-type: text/plain; charset=US-ASCII
Content-disposition: inline
Content-description: Attachment information.

The following section of this message contains a file attachment
prepared for transmission using the Internet MIME message format.
If you are using Pegasus Mail, or any another MIME-compliant system,
you should be able to save it or view it from within your mailer.
If you cannot, please ask your system administrator for assistance.

   ---- File information -----------
     File:  testres.src
     Date:  3 Mar 1998, 13:31
     Size:  5295 bytes.
     Type:  Unknown

--Message-Boundary-18041
Content-type: Application/Octet-stream; name="testres.src"; type=Unknown
Content-disposition: attachment; filename="testres.src"
Content-transfer-encoding: BASE64
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--Message-Boundary-18041--

---------- End of message ----------

From: "Tom Doan" <estima@estima.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Recursive Least Squares
Date: Thu, 18 Feb 1999 09:44:10 +0000
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Organization: Tom Doan
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
X-mailer: Pegasus Mail for Windows (v2.54) (via Mercury MTS (Bindery) v1.40)

The following procedure does recursive estimation producing a
series of recursive residuals for CUSUM style tests. The syntax
is very similar to a LINREG instruction, that is,

@RECRESID  depvar   start   end   resids
# list of explanatory variables

*
*  RECRESID computes recursive residuals for a linear model.  It is
*  somewhat fancier than the similarly named procedure in the RATS
*  manual, allowing start and end to be set by default.
*
*  Syntax:
*
*    RECRESID depvar start end resids
*    # explanatory variables
*
*    depvar = dependent variable
*    start  end = range of estimation
*    resids = (output) series of recursive residuals
*
*  Written, Sept. 1990 by Tom Doan
*  Updated, January , 1992
*
PROCEDURE RECRESID DEPVAR START END RESIDS
TYPE SERIES DEPVAR
TYPE INTEGER START END
TYPE SERIES *RESIDS
*
LOCAL EQUATION REGEQN
LOCAL INTEGER TIME STARTL ENDL
LOCAL INDEX REGLIST 
*
*  Get explanatory variables
*

ENTER(VARYING) REGLIST

*
*  Check that RESIDS is defined
*

IF .NOT.%DEFINED(RESIDS) {
   DISP 'Syntax: RECRESID  depvar  start  end  resids'
   RETURN
}

*
*  Determine range if START and END aren't set.
*

INQUIRE(REGRESSORLIST) STARTL>>START  ENDL>>END
# REGLIST DEPVAR

*
*  Run the regression, defining REGEQN
*

LINREG(DEFINE=REGEQN) DEPVAR STARTL ENDL
# REGLIST

SYSTEM(KALMAN) REGEQN
END(SYSTEM)
*
*  Estimate (perfect fit) over first NREG entries.  Kalman filter
*  over remainder.
*
ESTIMATE STARTL STARTL+%NREG-1
CLEAR RESIDS
DO TIME=STARTL+%NREG,ENDL
   KALMAN(RESIDS=RECURSIVE) RESIDS
END DO TIME
END

---------- End of message ----------

From: Irida Achthoven <achthovi@qed.econ.queensu.ca>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: garch standardized residuals
Date: Thu, 18 Feb 1999 15:44:14 -0800
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Organization: Queen's University
X-Mailer: Mozilla 3.04 (Win16; I) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit

Philippe PROTIN wrote:
> 
> Dear Irida,
> 
> Here is a procedure I wrote to test residuals : standardized ans
> squared standardized residuals (LB STAT AND BERA JARQUE TEST FOR
> NOMALITY...)
> 
> Hope it will help.
> 
>         \\\|||||///
>         \\  - -  //
>          (  @ @  )
> +------oOOo-(_)-oOOo---- +----------------------------------------+
> |                        | Philippe PROTIN                             |
> | Have a nice day !      | Ecole Supérieure des Affaires               


Hallo everybody
can anybody explain to me how to acces the residulas from garch
estimation. using the garch.prg
I need to acces them , so that I can apply the testres.src to it, for
ljungbox ect.
my mean egn is
R(t)=  B0 + B1*R(t-1) + E(t)
an example will be really appreciated
I really need the help urgently
CHEERS, IRIDA







> |                        | Tel: (33) 04.76.82.57.48                    |
> |                        | e-Mail: protin@esa.upmf-grenoble.fr         |
> +---------------Oooo-----+---------------------------------------------+
>         oooO   (   )
>        (   )    ) /
>         \ (    (_/
>          \_)

---------- End of message ----------

From: "Peijie Wang" <mcymswa@fs1.sm.umist.ac.uk>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: garch standardized residuals
Date: Thu, 18 Feb 1999 20:38:44 BST
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
X-mailer: Pegasus Mail for Windows (v2.54) (via Mercury MTS (Bindery) v1.40)

> From:          Irida Achthoven <achthovi@qed.econ.queensu.ca>
> To:            "RATS Discussion List" <RATS-L@efs.mq.edu.au>
> Subject:       garch standardized residuals
> Date:          Thu, 18 Feb 1999 07:02:03 -0800
> Reply-to:      "RATS Discussion List" <RATS-L@efs.mq.edu.au>
> Organization:  Queen's University

> I hope that somebody will reponds , because I really need the help
> 1. I have astimeted the garch(p,g) model. I know that the residuals and
> the  conditional covariance are resp sotred in the series U(T) and H(t)
> but they are not accesible.


The easiest (not neccesarily the best) way is to add two lines in 
garch.src (before END PROC GARCH and after Maximize ... ), such as: 

open copy res.dat
copy(dates,format=wks,org=obs) / u h

Then, after running the procedure, you can use a simple code to read  
u and h in and do your other tests. (specify a directory so you can 
find where file res.dat is). 

Alternatively, you can use a procedure garchuv.prg by incorporating 
it into your own programme. The problem is that you have to change 
something, though it is easy as the idea of the procedure is rather 
straightforward.

Hope these help.

Peijie Wang
UMIST


> Hi, 
> I am a new subsriber and I have search the archives to find the answer
> to my questions before posting here.
> unfortunately i have noticed that these questions are asked frequently,
> but there are no responses tothem. (at least not in the archives)
> 
> I hope that somebody will reponds , because I really need the help
> 1. I have astimeted the garch(p,g) model. I know that the residuals and
> the  conditional covariance are resp sotred in the series U(T) and H(t)
> but they are not accesible.
> HOW CAN I CALCULATE THE STANDARDIZED ERRORS
> HOW CAN I CALCULATE THE MOMENTS(KURTOSIS) OF THE STANDARD ERRORS.
> HOW CAN I CALCULATE THE LJUNG BOX TEST STAT. FOR SERIAL CORRELATION.
> 
> i HAVE TRIED TO USE THE GARCH SRC CODE, SINCE IT IS A VERY VALUABLE CODE
> BUT I HAVE RUN INTO A PROBLEM , AND DO NOT KNOW HOW TO SOLVE IT, THAT'S
> WHY I TRIED THE CONVENTIONAL WAY
> 
> HERE IS THE APRT AND THE DISPLAYED ERROR CODE,  I GOT WHEN IMPLEMENTING
> GARCH SRC
> 
> calendar(daily) 1994 1 4
> all 0 1999:1:20
> open data a:stock5.txt
> data(format=free,org=obs) / apple gate hewl micro intel m fr hola hol
> holp
> SET Lapple = log(apple)
> set Lgate =  log(gate)
> set Lhewl =  log(hewl)
> set Lmicro = log(micro)
> set Lintel = log(intel)
> set Dapple = Lapple -Lapple{1}
> set Dhewl = Lhewl- Lhewl{1}
> set Dgate  = Lgate- Lgate{1}
> set Dmicro = Lmicro- Lmicro{1}
> set Dintel = Lintel- Lintel{1}
> set y = dmicro
> source(noecho) c:\garch.src
> @garch(mod=arch,dist=ged,ar=1,p=1,q=0,iter=100,cc=0.08, qc1=0.08,bhhh)
> 94:01;05  98:06:22  y nresids  cvar
> 
> ## SX22. Expected Type SERIES, Got REAL Instead
> >>>>ev=levc,l1=lc,m=mc<<<<
> 
> I HOPE SOMEBODY CAN TELL ME HOW TO GO ABOUT THE 3 FIRST QUESTIONS OR
> TELL ME WHAT I DO WRONG IN THE ABOVE.
> 
> THANK YOU ,
> IRIDA
> END OF MESSAGE
> 

---------- End of message ----------

From: Norman Morin <m1njm00@frb.gov>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: garch standardized residuals
Date: Thu, 18 Feb 1999 16:08:20 -0500
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-version: 1.0
X-Mailer: exmh version 2.0.2.4 7/8/98 (via Mercury MTS (Bindery) v1.40)
Content-type: text/plain; charset=us-ascii


> > I hope that somebody will reponds , because I really need the help
> > 1. I have astimeted the garch(p,g) model. I know that the residuals and
> > the  conditional covariance are resp sotred in the series U(T) and H(t)
> > but they are not accesible.
> 
> 
> The easiest (not neccesarily the best) way is to add two lines in 
> garch.src (before END PROC GARCH and after Maximize ... ), such as: 
> 
> open copy res.dat
> copy(dates,format=wks,org=obs) / u h
> 

> > source(noecho) c:\garch.src
> > @garch(mod=arch,dist=ged,ar=1,p=1,q=0,iter=100,cc=0.08, qc1=0.08,bhhh)
> > 94:01;05  98:06:22  y nresids  cvar
 
Unless I am misunderstanding the question, from your code here, it looks
like you saved the normalized residuals to "nresids" and the 
conditional variances to "cvar".  If you want the "un-normalized"
residuals, you can just set
    resids = sqrt(cvar)*nresids


What's the "bhhh" line at the end of the procedure call?


Warmest regards,

Norm




-- 

                           Norman J. Morin 
                  nmorin@frb.gov or m1njm00@frb.gov         
          --------------------------------------------------
          Division of Research and Statistics * Mail Stop 82
           Board of Governors of the Federal Reserve System 
               Washington, D.C. 20551 * (202) 452-2476
          --------------------------------------------------



---------- End of message ----------

From: Norman Morin <m1njm00@frb.gov>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: GARCH
Date: Thu, 18 Feb 1999 16:16:58 -0500
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-version: 1.0
X-Mailer: exmh version 2.0.2.4 7/8/98 (via Mercury MTS (Bindery) v1.40)
Content-type: text/plain; charset=us-ascii

Hello,

I neglected to define some numbers used in GARCH.SRC and 
GARCHN.SRC as local variables.  This has caused some people
problems when using the programs.  Sorry about that!  

So, to remedy this, in GARCH.SRC and GARCHN.SRC, about 510 lines
down into the program, if you insert the following lines of code
with the rest of the LOCAL variable statements, it might make
some problems disappear.


LOCAL REAL   c a1 a2 a3 a4 a5 a6 a7 a8 a9 a10 a11 a12 a13
LOCAL REAL   b1 b2 b3 b4 b5 b6 b7 b8 b9 b10 b11 b12 b13 b14 $
             b15 b16 b17 b18 b19 b20 b21 b22 b23 b24 b25
LOCAL REAL   vx1 vx2 vx3 vx4 vx5 vx6 vx7 vx8 vx9 vx10 vx11 $
             vx12 vx13 p1 p2 p3 p4 q1 q2 q3 q4 lev D M L1


Sorry, again.


Warmest regards,

Norm
-- 

                           Norman J. Morin 
                  nmorin@frb.gov or m1njm00@frb.gov         
          --------------------------------------------------
          Division of Research and Statistics * Mail Stop 82
           Board of Governors of the Federal Reserve System 
               Washington, D.C. 20551 * (202) 452-2476
          --------------------------------------------------



---------- End of message ----------

From: Irida Achthoven <achthovi@qed.econ.queensu.ca>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: [Fwd: garch.src.and winrats crash]
Date: Fri, 19 Feb 1999 15:00:48 -0800
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Organization: Queen's University
X-Mailer: Mozilla 3.04 (Win16; I) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: message/rfc822
Content-Transfer-Encoding: 7bit

Message-ID: <36CDCD54.C2F@qed.econ.queensu.ca>
Date: Fri, 19 Feb 1999 12:45:08 -0800
From: Irida Achthoven <achthovi@qed.econ.queensu.ca>
Organization: Queen's University
X-Mailer: Mozilla 3.04 (Win16; I)
MIME-Version: 1.0
To: MAISER@EFS.MQ.EDU.AU
Subject: garch.src.and winrats crash
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

hallo everybody
and Mr.Tom maycock
I finally got my parch.src running. but now winrats keeps crashing on
me.
I have a 386 and version 4.3
can anybody offer me a solution.
I really need to move on with my research
thanks ,
wishing everybody a nice weekend
ps i hope mr maycock will respond



---------- End of message ----------

From: "Wai Lee" <lee_wai@jpmorgan.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: [Fwd: garch.src.and winrats crash]
Date: Fri, 19 Feb 1999 15:54:07 -0500
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Mime-Version: 1.0
Content-type: text/plain; charset=us-ascii
X-Mailer: Mercury MTS (Bindery) v1.40

I certainly understand your situations.  The crashing problem of RATS has
been haunting me for years.  I just got updated to WinRATS-32 4.31.  It
still crashes when I highlight some parts of my program with mouse.  This
is a known problem.  Other than this, I haven't experienced any other crash
problems that I had with version 4.21.  Since RATS is a good software, I am
willing to live with the crashes.  Just save your program frequently.

W. Lee





achthovi @ qed.econ.queensu.ca on 02/19/99 06:00:48 PM

Please respond to RATS-L@efs.mq.edu.au

To:   RATS-L @ efs.mq.edu.au
cc:   (bcc: Wai Lee)
Subject:  [Fwd: garch.src.and winrats crash]





Message-ID: <36CDCD54.C2F@qed.econ.queensu.ca>
Date: Fri, 19 Feb 1999 12:45:08 -0800
From: Irida Achthoven <achthovi@qed.econ.queensu.ca>
Organization: Queen's University
X-Mailer: Mozilla 3.04 (Win16; I)
MIME-Version: 1.0
To: MAISER@EFS.MQ.EDU.AU
Subject: garch.src.and winrats crash
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit



hallo everybody
and Mr.Tom maycock
I finally got my parch.src running. but now winrats keeps crashing on
me.
I have a 386 and version 4.3
can anybody offer me a solution.
I really need to move on with my research
thanks ,
wishing everybody a nice weekend
ps i hope mr maycock will respond









---------- End of message ----------

From: khl2288@UTARLG.UTA.EDU
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Concerning TESTRES.SRC: garch standardized residuals
Date: Fri, 19 Feb 1999 16:15:58 -0600 (CST)
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-version: 1.0
Content-type: TEXT/PLAIN; charset=US-ASCII
X-Mailer: Mercury MTS (Bindery) v1.40

Hi, RATS users and Philippe Protin.
I tried the Testres.src, but had a syntax error message:
## SX11. Identifier NOBS is not recognizabe.....
>>>>te lbstats = nobs*(<<<<

I found this error occured in the LB procedure. The file I downloaded
includes two compute statements around the end of the code for the LB
procedure;
if (spanl.lt.lags) {
compute count = spanl
.......
    else
    compute count = count+spanl
    }
   }
compute lbstats = nobs*(nobs+2)*csum(lags)
}
compute lbstats = nobs*(nobs+2)*csum(lags)
compute lbsig = ........

I tried a couple of things to find the cause of error, and found that it
occoured at the second Compute statement for lbstats. I ran the program
again after deleting } and the second Compute statement for lbstats, and
got no error. Did I do a right correction?
Please reply me.

Thank you so much.
 


Kyong H. Lee
Dept. of Economics
University of Texas at Arlington


---------- End of message ----------

From: khl2288@UTARLG.UTA.EDU
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: testres.src
Date: Fri, 19 Feb 1999 18:00:32 -0600 (CST)
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-version: 1.0
Content-type: TEXT/PLAIN; charset=US-ASCII
X-Mailer: Mercury MTS (Bindery) v1.40

Hi, RATS users and Philippe Protin.
I got two problems in running the testres.src.
First, in compling the testres.src, I got a syntax error message:
## SX11. Identifier NOBS is not recognizable.....
>>>>te lbstats = nobs*(<<<<

So, I tried a couple of things and found that the error occured at the
second Compute statement for lbstats in LB procedure code. The program
file I downloaded includes:
procedure LB depvar start end
........
.......
if (spanl.lt.lags) {
........
    if (count.lt.lags) {
........
       else
       compute count = count+spanl
       }
    }
compute lbstats = nobs*(nobs+2)*csum(lags)
    }
compute lbstats = nobs*(nobs+2)*csum(lags).

So, I deleted the second Compute statement and }(which looks to be
redundant) above the Compute statement, and could compile the testres.src
and got the first three test values; BERA, LB, and ARCH.
Did I make a right correction?

Second problem involves in the ANALRES procedure.
A portion of my program is:
source(noecho) c:\testres.src
@bera y
@lb y
@archtest y
@analres y h

I got an error message:
@analres <<<<
## CP18. ANALRES is not the name of a procedure. (Did you forget to
Source?)

Please help me out.

Thanks,
 


Kyong H. Lee
Dept. of Economics
University of Texas at Arlington


---------- End of message ----------

From: AMMAR <abdelma@ksu.edu.sa>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: garch standardized residuals
Date: Fri, 19 Feb 1999 20:20:05 +0300
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
X-Mailer: Mozilla 4.04 [en] (Win95; I) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

> > > source(noecho) c:\garch.src
> > > @garch(mod=arch,dist=ged,ar=1,p=1,q=0,iter=100,cc=0.08, qc1=0.08,bhhh)
> > > 94:01;05  98:06:22  y nresids  cvar
>
>
> What's the "bhhh" line at the end of the procedure call?
>

Berndt, Hall, Hall and Hausman cov mtx.
Ammar


---------- End of message ----------

From: rawia atef mokhtar <rawiam@aucegypt.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Neural Networks
Date: Sat, 20 Feb 1999 14:22:22 +0200
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Organization: American University in Cairo
X-Mailer: Mozilla 4.02 [en] (Win95; I) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Dear  RATS users,
I am doing a thesis on exchange rate forecasting  and I am using neural
networks to forecast
Canadian $ / DM. My problem is that after I train the networks using
NNLEARN , I fail
to save the memory vetor to retrieve it whenever I want to use it again.
I write the following
to save the vector:
open copy mv.txt
copy(org=variable, format=free)
write (unit=copy) mv

And I write the following to retrieve the vector again:
declare vector mv
open data  mv.txt
read(varying) mv

At very little times I managed to save the vector but I was never able
to read it again to use
it again in nnlearn and then in nntest.. Can anyone help me in that?

Thanks

Rawia Atef Mokhtar
American university in Cairo
Egypt


---------- End of message ----------

From: Simon.Van-Norden@hec.ca
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Neural Networks
Date: Sat, 20 Feb 1999 23:13:10 -0500
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Organization: Ecole des Hautes Etudes Commerciales
X-Mailer: Mozilla 4.5 [en] (Win95; U) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit

Rawia;

Did you try deleting the COPY statement from your program when you write the
vector?  I don't think it is required.

SvN

rawia atef mokhtar wrote:
> 
> Dear  RATS users,
> I am doing a thesis on exchange rate forecasting  and I am using neural
> networks to forecast
> Canadian $ / DM. My problem is that after I train the networks using
> NNLEARN , I fail
> to save the memory vetor to retrieve it whenever I want to use it again.
> I write the following
> to save the vector:
> open copy mv.txt
> copy(org=variable, format=free)
> write (unit=copy) mv
> 
> And I write the following to retrieve the vector again:
> declare vector mv
> open data  mv.txt
> read(varying) mv
> 
> At very little times I managed to save the vector but I was never able
> to read it again to use
> it again in nnlearn and then in nntest.. Can anyone help me in that?
> 
> Thanks
> 
> Rawia Atef Mokhtar
> American university in Cairo
> Egypt

-- 
Simon van Norden			Professeur invité
E-mail:  simon.van-norden@hec.ca   or    svn@alum.mit.edu 
Home Page:       http://www.hec.ca/pages/simon.van-norden
(514)340-6781 or fax (514)340-5632 or  cell (514)993-0466
Service de l'enseignement de la finance, École des H.E.C.
3000 Cote-Sainte-Catherine, Montreal QC, CANADA   H3T 2A7

---------- End of message ----------

From: rawia atef mokhtar <rawiam@aucegypt.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: RE:NEURAL NETWORKS
Date: Sun, 21 Feb 1999 12:01:01 +0200
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Organization: American University in Cairo
X-Mailer: Mozilla 4.02 [en] (Win95; I) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Hi Hardy,
Thanks for the reply.. Yes I  heard of  Neuroshell but I have already
started working
in RATS and the computer is training the network since 5 days ago and I
don't want
in this moment to switch to other program for this small problem.. But I
promise
I will look at it..
I am using the variables in the monetary model of exchange rate
determination to forecast
the exchange rate. These are money,income and interest rate
differentials between two
countries.. What did you use?


---------- End of message ----------

From: rawia atef mokhtar <rawiam@aucegypt.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: re:neural networks
Date: Sun, 21 Feb 1999 12:04:10 +0200
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Organization: American University in Cairo
X-Mailer: Mozilla 4.02 [en] (Win95; I) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Svn ;
Thanks. I will try it once I stop trainig the networks and see if it
works
rawia



---------- End of message ----------

From: khl2288@UTARLG.UTA.EDU
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: LB test with a large no. of obs.
Date: Mon, 22 Feb 1999 00:07:27 -0600 (CST)
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-version: 1.0
Content-type: TEXT/PLAIN; charset=US-ASCII
X-Mailer: Mercury MTS (Bindery) v1.40

Hi, RATS users.
I'm trying to estimate GARCH-M model using daily data. The number of
observations are about 2800. When I test the serial correlation of the raw
data and the (standardized)residuals from the model, the null of no
autocorrelation is rejected with very low significance level(up to lag 12) 
even though the actual serial correlation coefficients are very small. The
largest correlation coefficients are around 0.05. Is it because of the
large number of observation?
I'll greatly appreciate for any comment.


Kyong H. Lee
Dept. of Economics
University of Texas at Arlington


---------- End of message ----------

From: "Estima" <estima@estima.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: [Fwd: garch.src.and winrats crash]
Date: Mon, 22 Feb 1999 14:12:19 -0600
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
X-mailer: Pegasus Mail for Win32 (v2.54) (via Mercury MTS (Bindery) v1.40)

> 
> hallo everybody
> and Mr.Tom maycock
> I finally got my parch.src running. but now winrats keeps crashing on
> me.
> I have a 386 and version 4.3
> can anybody offer me a solution.
> I really need to move on with my research
> thanks ,
> wishing everybody a nice weekend
> ps i hope mr maycock will respond
> 

For help with this, please contact us directly at 
"estima@estima.com". I'll need to know your RATS serial number, and 
as much information as you can provide about where and how the 
program crashes.

Sincerely,
Tom Maycock
Estima


------------------------------------------------------------
|   Estima                    |  Sales:   (800) 822-8038   |
|   P.O. Box 1818             |  Support: (847) 864-1910   |
|   Evanston, IL 60204-1818   |  Fax:     (847) 864-6221   |
|   USA                       |  estima@estima.com         |
|                             |  http://www.estima.com     |
------------------------------------------------------------

---------- End of message ----------

From: "Estima" <estima@estima.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Neural Networks
Date: Mon, 22 Feb 1999 14:12:19 -0600
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
X-mailer: Pegasus Mail for Win32 (v2.54) (via Mercury MTS (Bindery) v1.40)

You need to omit the COPY command--that's going to write out all the 
data series currently in memory to the file. You want the memory 
vector to go into it's own file. Also, make sure that you don't put 
any blank spaces between the instruction name, such as "write" and 
the option field "(unit=copy)". For example, try the following:

open copy mv.txt
write(unit=copy) mv

If you want to be able to read the file back in the same RATS session 
(i.e. without first having quit and restarted RATS), you need to be 
sure to close the file for writing before re-opening it for reading. 
Thus I would generally recommend including a CLOSE COPY command after 
the above:

open copy mv.txt
write(unit=copy) mv
close copy

You can then do:

declare vector mv
open data  mv.txt
read(varying) mv

to read in the vector.

Sincerely,
Tom Maycock
Estima
 

------------------------------------------------------------
|   Estima                    |  Sales:   (800) 822-8038   |
|   P.O. Box 1818             |  Support: (847) 864-1910   |
|   Evanston, IL 60204-1818   |  Fax:     (847) 864-6221   |
|   USA                       |  estima@estima.com         |
|                             |  http://www.estima.com     |
------------------------------------------------------------

---------- End of message ----------

From: "mcymswa" <mcymswa@fs1.sm.umist.ac.uk>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: testres.src
Date: Mon, 22 Feb 1999 22:06:52 BST
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
X-mailer: Pegasus Mail for Windows (v2.54) (via Mercury MTS (Bindery) v1.40)

> From:          khl2288@UTARLG.UTA.EDU
> To:            "RATS Discussion List" <RATS-L@efs.mq.edu.au>
> Subject:       testres.src
> Date:          Fri, 19 Feb 1999 18:00:32 -0600 (CST)
> Reply-to:      "RATS Discussion List" <RATS-L@efs.mq.edu.au>

> 
> Second problem involves in the ANALRES procedure.
> A portion of my program is:
> source(noecho) c:\testres.src
> @bera y
> @lb y
> @archtest y
> @analres y h
> 
> I got an error message:
> @analres <<<<
> ## CP18. ANALRES is not the name of a procedure. (Did you forget to
> Source?)
> 
Perhaps you don't mind checking whether all involved procedures are 
under the right directories and setting your RATS working directory 
accordingly.

Peijie

> Please help me out.
> 
> Thanks,
>  
> 
> 
> Kyong H. Lee
> Dept. of Economics
> University of Texas at Arlington
> 
> 

---------- End of message ----------