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 UAA263922
	for <BAUM@monet.bc.edu>; Tue, 3 Mar 1998 20:27:54 -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 UAA153980
	for BAUM@MAIL1.BC.EDU; Tue, 3 Mar 1998 20:27:53 -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 UAA151638
	for <baum@bc.edu>; Tue, 3 Mar 1998 20:27:39 -0500
Received: from efs1.efs.mq.edu.au (efs1.efs.mq.edu.au [137.111.64.8])
	by sunb.ocs.mq.edu.au (8.8.5/8.8.5) with ESMTP id MAA18497
	for <baum@bc.edu>; Wed, 4 Mar 1998 12:27:34 +1100 (EST)
Received: from EFS1/SpoolDir by efs1.efs.mq.edu.au (Mercury 1.21);
    4 Mar 98 12:27:13 GMT+1000
Received: from SpoolDir by EFS1 (Mercury 1.30); 4 Mar 98 12:27:12 GMT+1000
To: baum@bc.edu
Date: Wed, 4 Mar 98 12:27:12 GMT+1000
Subject: Re: 
Message-ID: <31C865F2C91@efs1.efs.mq.edu.au>

From: John Cline <jcline@glinx.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: RATS disscusion
Date: Sat, 31 Jan 1998 14:44:36 -0400
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) v1.30

>> Hello John,
>
>Again, I will make no promises as to when a new version of MacRATS
>will be available.  We have several options available.  I can refund
>the $100 you paid for a Mac version of CATS.  You can upgrade your
>current version of RATS to a Windows application and have CATS for
>a PC or you can continue to wait.  The choice, as in the past, is
>yours. I wish I could give you more information but obviously, I have
>no idea when.  Let me know how you wish to handle this.
>
>Regards,
>Susan Bessonny
>

Please send me the refund and remove me from your discussion list.  Job I
am not.

Regretfully,

John Cline

Amaranth And Associates &
Amaranth Co-operative Enterprises Ltd.
PO Box 448
Wolfville, Nova Scotia,  B0P 1X0
1-902-542-4002 ( Fax & Voice )
Email:  jcline@glinx.com
Homepage:  http://www.glinx.com/users/jcline/
Worker Owned

Environmental Products, Education, Sustainable Development And Consulting
For A Conserver Society



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

From: Felix.Maag@sbf.unisg.ch
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: White/Newey-West
Date: Mon, 2 Feb 1998 10:43:34 +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
Content-type: text/plain; charset=us-ascii
X-Mailer: Mercury MTS (Bindery) v1.30

Dear Rats-user.

I want to correct my regressions for heteroscedasticity. To do so I would
like to implement White's correction for heteroscedasticity (White[1980])
or Newey-West's correction for heteroscedasticity and serial correlation
(Newey/West [1987]).
Can sombody tell me how to implement them with Rats?

Thanks for help.

Felix Maag



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

From: "Philippe PROTIN" <PROTIN@esa.upmf-grenoble.fr>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: White/Newey-West
Date: Mon, 2 Feb 1998 11:19:44 +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=US-ASCII
Content-transfer-encoding: 7BIT
X-mailer: Pegasus Mail for Windows (v2.54) (via Mercury MTS (Bindery) v1.30)

Dear Felix,

To implement White's correction for heteroskedasticity in your 
regression, you can use the robusterrors, damp and lags options in 
the linreg procedure. Lags is required if you need to correct for 
heteroskedasticity and autocorrelation. I think the Newey/West 
correction need to put DAMP=1.0. See the manual for more details.

Philippe
*********************
Philippe PROTIN
ESA-CERAG
BP 47X
38040 GRENOBLE CEDEX 09
FRANCE

04.76.82.57.48.
protin@esa.upmf-grenoble.fr

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

From: "Estima" <estima@estima.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: RATS disscusion
Date: Mon, 2 Feb 1998 09:32:30 -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
Content-transfer-encoding: 7BIT
X-mailer: Pegasus Mail for Win32 (v2.54) (via Mercury MTS (Bindery) v1.30)

> 
> Please send me the refund and remove me from your discussion list.  Job I
> am not.
> 

John:

I forwarded your reply to Sue. If you want off the RATS mailing list, 
you need to do that yourself--using the instructions you were sent 
when you signed up (these instructions are also available on our Web 
site).

Tom Maycock
Estima

--
+-----------------------------+-----------------------------------------+
| Estima                      |                                         |
| P.O. Box 1818               |  Voice: (847) 864-8772                  |
| Evanston, IL 60204-1818     |  Fax:   (847) 864-6221                  | 
| U.S.A                       |  BBS:   (847) 864-8816                  |
| e-mail: estima@estima.com   |  CompuServe: 73140,2202                 |
|-----------------------------------------------------------------------|
| Web Site:  http://www.estima.com                                      |
| RATS Internet Mailing List: New members can join by sending e-mail to |
|  MAISER@EFS.MQ.EDU.AU with the message: SUBSCRIBE  RATS-L             |
+-----------------------------------------------------------------------+

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

From: rkraeussl@wiwi.uni-bielefeld.de
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Swamy's Random Coefficient Estimator 
Date: Mon, 2 Feb 1998 17:20:38 GMT+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: Fak.WIWI, Uni Bielefeld
X-mailer: Pegasus Mail for Windows (v2.54) (via Mercury MTS (Bindery) v1.30)

Hello RATS User,

I'm looking for a RATS procedure which implements Swamy's random 
coefficient estimator, which is described in his article 

"Linear Prediction and Estimation Methods for Regression Methods with 
Stationary Stochastic Coefficients" in Journal of Econometrics 
(1980), 

resp. Chang's article "Efficient Computation of  Stochastic 
Coefficient Models" in Amman's book "Computational Economics and 
Econometrics" (1992).

Thanks in advance, Roman Kraeussl

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

From: Hakan Berument <berument@Bilkent.EDU.TR>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Multivariate Generalized Error Distribution
Date: Mon, 02 Feb 1998 18:26:06 +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>
X-Mailer: Mozilla 4.03 [en] (Win95; I) (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

I am estimating a multivariate GARCH in Means model. When the Normality
of the error term is assumed wrongly, standartd errors turns to be
wrong.  However, Harvey and Nelson  proposes using General Error
Distribution for the maximization.  The procedure is available for the
univariate case in @GARCH procedure in RATS.  I wonder  whether anybody
know how how to write the multivariate version of the Gerealized Error
Distribution for the Mutivariate GARCH in Means procedure.

Thanks in advance.

Hakan Berument
Department of Economics
Bilkent University
Ankara Turkey


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

From: Michael Hanson <mhanson@econ.lsa.umich.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: RATS disscusion
Date: Mon, 2 Feb 1998 16:27:43 -0500 (EST)
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) v1.30

	I have a follow-up question regarding opening FAME databases with
RATS on Unix.  While the command "open data fame_db" (as per the man page) 
does not return an error, I have yet to figure out how to actually DO
anything with the data -- I cannot seem to access it.  Is there a
subsequent DATA(FORMAT=FAME... or the like which I should be using?  We
don't seem to have received any additional documentation (beyond the man
page blurb) for dealing with FAME databases when we upgraded to v4.3. TIA. 

======================================================================
Michael S. Hanson				     mhanson@umich.edu
Department of Economics		http://www.econ.lsa.umich.edu/~mhanson
University of Michigan


On Fri, 30 Jan 1998 Pierre.SLADDEN@DG2.cec.be wrote:

> With WinRATS 4.3 (32bits version) comes RATSData (16bits component->some problems) who permits the data import from different origins (TSP, Lotus, ...).
> The Unix version of RATS 4.3 has an interface to access directly FAME databases (update mode available).
> 
> Regards,
> 
> Pierre Sladden
> European Commission
> DG2 - Brussels
> 
> 




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

From: Norman Morin <m1njm00@Frb.GOV>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: RATS disscusion 
Date: Mon, 02 Feb 1998 16:56:57 -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: exmh version 1.6.9.4 5/23/97 (via Mercury MTS (Bindery) v1.30)
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii

> 	I have a follow-up question regarding opening FAME databases with
> RATS on Unix.  While the command "open data fame_db" (as per the man page) 
> does not return an error, I have yet to figure out how to actually DO
> anything with the data -- I cannot seem to access it.  Is there a
> subsequent DATA(FORMAT=FAME... or the like which I should be using? 


At the Fed, the variables in FAME always have the form, say, gdp.q.
So one has to create a text file with two columns, the left column being
the name you want to assign the variable for RATS (since RATS thinks
periods are for separating logical operators) and the right column being
the FAME name.  I usually just replace the "." with a "_" whenever
they occur (e.g., gdp.q becomes gdp_q).

Call the two-column text file famenames.txt, for eaxmple.  And suppose
the FAME database is fame.db.

Then in RATS you open the names file:

	OPEN FAME famenames.txt

Then open the FAME database:

	OPEN DATA fame.db

Then read in the series you want, using the RATS names you have in the
names file
	DATA(format=fame) / gdp_q cons_q


And then you're in the clear.  If the variable names in fame.db are ok for
RATS, you won't need the names file.  Unlike RATS db's or wks files, you
have to list the series you want, instead of reading in the entire thing.
I think this is because FAME db's often have tens of thousands of series. 
 

I hope this helps. 


Norm

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



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

From: "Estima" <estima@estima.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: RATS disscusion
Date: Mon, 2 Feb 1998 16:20:30 -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
Content-transfer-encoding: 7BIT
X-mailer: Pegasus Mail for Win32 (v2.54) (via Mercury MTS (Bindery) v1.30)


> 	I have a follow-up question regarding opening FAME databases with
> RATS on Unix.  While the command "open data fame_db" (as per the man page) 
> does not return an error, I have yet to figure out how to actually DO
> anything with the data -- I cannot seem to access it.  Is there a
> subsequent DATA(FORMAT=FAME... or the like which I should be using?  We
> don't seem to have received any additional documentation (beyond the man
> page blurb) for dealing with FAME databases when we upgraded to v4.3. TIA. 
> 

Yes, the command is:

DATA(FORMAT=FAME)

If you have the FAME interface option, you would have received some 
manual pages describing how to set up and use the FAME interface, so 
you may want to try and locate those.

Sincerely,
Tom Maycock
Estima
 
--
+-----------------------------+-----------------------------------------+
| Estima                      |                                         |
| P.O. Box 1818               |  Voice: (847) 864-8772                  |
| Evanston, IL 60204-1818     |  Fax:   (847) 864-6221                  | 
| U.S.A                       |  BBS:   (847) 864-8816                  |
| e-mail: estima@estima.com   |  CompuServe: 73140,2202                 |
|-----------------------------------------------------------------------|
| Web Site:  http://www.estima.com                                      |
| RATS Internet Mailing List: New members can join by sending e-mail to |
|  MAISER@EFS.MQ.EDU.AU with the message: SUBSCRIBE  RATS-L             |
+-----------------------------------------------------------------------+

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

From: John Cline <jcline@glinx.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: RATS disscusion
Date: Mon, 2 Feb 1998 21:25:56 -0400
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) v1.30

>>
>> Please send me the refund and remove me from your discussion list.  Job I
>> am not.
>>
>
>John:
>
>I forwarded your reply to Sue. If you want off the RATS mailing list,
>you need to do that yourself--using the instructions you were sent
>when you signed up (these instructions are also available on our Web
>site).
>
>Tom Maycock
>Estima


Thanks and I will watch for a remove as it occurs next on the list.  I can
honestly say I am disappointed and hope my refund occurs quickly.

John Cline

Amaranth And Associates &
Amaranth Co-operative Enterprises Ltd.
PO Box 448
Wolfville, Nova Scotia,  B0P 1X0
1-902-542-4002 ( Fax & Voice )
Email:  jcline@glinx.com
Homepage:  http://www.glinx.com/users/jcline/
Worker Owned

Environmental Products, Education, Sustainable Development And Consulting
For A Conserver Society



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

From: Pierre.SLADDEN@DG2.cec.be
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re(2): RATS disscusion
Date: Tue, 3 Feb 1998 09:54:11 +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
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mercury MTS (Bindery) v1.30

Michael S. Hanson (mhanson@umich.edu) wrote:

>         I have a follow-up question regarding opening FAME databases with
> RATS on Unix.  While the command "open data fame_db" (as per the man page) 
> does not return an error, I have yet to figure out how to actually DO
> anything with the data -- I cannot seem to access it.  Is there a
> subsequent DATA(FORMAT=FAME... or the like which I should be using?  We
> don't seem to have received any additional documentation (beyond the man
> page blurb) for dealing with FAME databases when we upgraded to v4.3. TIA. 


Normally, the RATS v4.3 on Unix is delivered with a library named 'fame_demo' containing a lot of examples. In any case be careful when you use the 'fame_db' variable, it must contain the complete path of your file. Then use a simple 'data(format=fame)

 Pierre Sladden
 European Commission
 DG2 - Brussels

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

From: Jongchurl Kim <kim.317@osu.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: RATS disscusion
Date: Tue, 3 Feb 1998 06:52:05 -0500 (EST)
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: Windows Eudora Pro Version 2.1.2 (via Mercury MTS (Bindery) v1.30)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"

I wonder I should get this kind of personal message.

At 09:25 PM 2/2/98 -0400, you wrote:
>>>
>>> Please send me the refund and remove me from your discussion list.  Job I
>>> am not.
>>>
>>
>>John:
>>
>>I forwarded your reply to Sue. If you want off the RATS mailing list,
>>you need to do that yourself--using the instructions you were sent
>>when you signed up (these instructions are also available on our Web
>>site).
>>
>>Tom Maycock
>>Estima
>
>
>Thanks and I will watch for a remove as it occurs next on the list.  I can
>honestly say I am disappointed and hope my refund occurs quickly.
>
>John Cline
>
>Amaranth And Associates &
>Amaranth Co-operative Enterprises Ltd.
>PO Box 448
>Wolfville, Nova Scotia,  B0P 1X0
>1-902-542-4002 ( Fax & Voice )
>Email:  jcline@glinx.com
>Homepage:  http://www.glinx.com/users/jcline/
>Worker Owned
>
>Environmental Products, Education, Sustainable Development And Consulting
>For A Conserver Society
>
>
>
>

   =============================================================================
   Jongchurl Kim                                     Department of Economics
   kim.317@osu.edu                                   The Ohio State University
   Office (614)292-5765                              311 Arps Hall
          (614)292-3906(FAX)                         1945 North High Street
                                                     Columbus,Ohio 43210-1172


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

From: WOO KAI YIN <hkwoo@hkabc.net>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Inquiry about DISCO
Date: Wed, 04 Feb 1998 01:20:12 +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>
X-Mailer: Mozilla 3.0 (Win95; I) (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Dear Sir,

I'm writing a paper in which I am required to run Johansen's LR test.
However, in my model, intervention dummies are specified. I know that
the program DISCO (which can be used in CATS in RATS) is useful to
simulate critical values. So, could you let me know how and where can I
get this software, please? 

Thank you.

Regards,

Woo

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

From: WOO KAI YIN <hkwoo@hkabc.net>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Inquiry about file in MircoTSP
Date: Wed, 04 Feb 1998 01:26:26 +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>
X-Mailer: Mozilla 3.0 (Win95; I) (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Dear Sir,

I ask whether I can read files from MircoTSP from RATSDATA. If I can,
could you show me the commands? Thank you for your help!

Regards,

K.Y.Wu

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

From: jtebeka@oddo.fr
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: =?iso-8859-1?Q?R=E9f._:_Inquiry_about_DISCO?=
Date: Tue, 3 Feb 1998 18:33:51 +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
Content-type: text/plain; charset=us-ascii
X-Mailer: Mercury MTS (Bindery) v1.30




I think it is on Johansen's Home page. I do not remember the exact address
but you can use a search program on internet.



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

From: WOO KAI YIN <hkwoo@hkabc.net>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: inquiry about Disco
Date: Wed, 04 Feb 1998 02:18:39 +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>
X-Mailer: Mozilla 3.0 (Win95; I) (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Dear Sir,

In Johansen's home page, I found the "Download Disco" command. However,
it is so called zip file. what is this  file? Where is the Disco
download, in RATS, or elsewhere? What are the procedure to simulate the
critical values by use of this program? Thank you for your instrucitons.

Regards,

Wu

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

From: "=?ISO-8859-1?Q?=92=86=92J=95=FC=8F=BA?=" <nakatani@obihiro.ac.jp>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: =?ISO-8859-1?Q?RE:_R=E9f._:_Inquiry_about_DISCO?=
Date: Wed, 4 Feb 1998 03:32:02 +0900
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 4.70.1161 (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

Dear colleagues,

> I think it is on Johansen's Home page. I do not remember the exact
address
> but you can use a search program on internet.
> 

Although I am not sure the procedure can be done by DisCO, Johansen's Home
page address is http://www.math.ku.dk/~sjo/, where you will find a link to
download DisCO.  


//////////////////////////////////////////
Nakatani Tomoaki  nakatani@obihitro.ac.jp
Assistant Professor, Obihiro Univ. of Agriculture
Obihiro-shi Inada-cho, 080, Hokkaido, Japan
Phone/FAX:+81-155-49-5447 
//////////////////////////////////////////


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

From: "Estima" <estima@estima.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: inquiry about Disco
Date: Tue, 3 Feb 1998 12:35:23 -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: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
X-mailer: Pegasus Mail for Win32 (v2.54) (via Mercury MTS (Bindery) v1.30)

> 
> In Johansen's home page, I found the "Download Disco" command. However,
> it is so called zip file. what is this  file? Where is the Disco
> download, in RATS, or elsewhere? What are the procedure to simulate the
> critical values by use of this program? Thank you for your instrucitons.
> 

A Zip file is a file that has been compressed using the utility 
PK-Zip (or a compatible program). Once you have downloaded the file, 
you will need to use PKUNZIP or a compatible program to unzip the 
file. If you don't have PKUNZIP, you can download a copy from our Web 
site (www.estima.com).

Tom Maycock
Estima
 
--
+-----------------------------+-----------------------------------------+
| Estima                      |                                         |
| P.O. Box 1818               |  Voice: (847) 864-8772                  |
| Evanston, IL 60204-1818     |  Fax:   (847) 864-6221                  | 
| U.S.A                       |  BBS:   (847) 864-8816                  |
| e-mail: estima@estima.com   |  CompuServe: 73140,2202                 |
|-----------------------------------------------------------------------|
| Web Site:  http://www.estima.com                                      |
| RATS Internet Mailing List: New members can join by sending e-mail to |
|  MAISER@EFS.MQ.EDU.AU with the message: SUBSCRIBE  RATS-L             |
+-----------------------------------------------------------------------+

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

From: Pierre.SLADDEN@DG2.cec.be
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Inquiry about file in MircoTSP
Date: Wed, 4 Feb 1998 08:38:19 +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
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mercury MTS (Bindery) v1.30

Hello,

Extract from the RATSDATA Help :

"Import from MicroTSP
"
"This Import Menu operation imports data from MicroTSP .DB format files into the current data file.  A "Data File Window must be active 
"to perform this operation.
"
"Because MicroTSP data bases actually are stored as a separate file for each series (with .DB "extensions), this conversion process is 
"rather different from the others.  All .DB files on the current default directory are converted.  The "labeling and date information is 
"obtained from each file.
"
"If you only want to import some of the .DB files from a given directory, you'll need to copy or move "the desired files to a separate 
"directory and make that the default directory using the Directory... operation on the File Menu.  "Then import the data.

Pierre Sladden






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

From: jtebeka@oddo.fr
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: statistics on daily data
Date: Wed, 4 Feb 1998 09:42:12 +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
Content-type: text/plain; charset=us-ascii
X-Mailer: Mercury MTS (Bindery) v1.30




Dear all, i have daily data and i would like to calculate statistics
(means, variance, correlations,...) on a monthly basis. I mean i would like
to have the mean, variance, etc for the january month, february month,...
Is there anyone who can help me on that subject.
Moreover, i would like to test the stability of a correlation matrix over
time, do you know any test related to that subject.
Thanks for your help.
Jacques.



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

From: <anpe08@handelsbanken.se>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Data storing problem
Date: Wed, 4 Feb 1998 17:45:44 +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
Content-type: text/plain; charset=us-ascii
X-Mailer: Mercury MTS (Bindery) v1.30





Hi!

I have problem writing a matrix to an output file.
The size is 25 times 63 (rows*columns).
Write scrambles the output to only five columns wide and
display with hold can only handle 19 columns?!

Are there any useful tricks I could try?
Platform: WinRats

Regards,
Anders
anpe08@handelsbanken.se





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

From: WOO KAI YIN <hkwoo@hkabc.net>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: inquiry about Disco
Date: Thu, 05 Feb 1998 02:31:45 +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>
X-Mailer: Mozilla 3.0 (Win95; I) (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Estima wrote:
> 
> >
> > In Johansen's home page, I found the "Download Disco" command. However,
> > it is so called zip file. what is this  file? Where is the Disco
> > download, in RATS, or elsewhere? What are the procedure to simulate the
> > critical values by use of this program? Thank you for your instrucitons.
> >
> 
> A Zip file is a file that has been compressed using the utility
> PK-Zip (or a compatible program). Once you have downloaded the file,
> you will need to use PKUNZIP or a compatible program to unzip the
> file. If you don't have PKUNZIP, you can download a copy from our Web
> site (www.estima.com).
> 
> Tom Maycock
> Estima
> 
> --
> +-----------------------------+-----------------------------------------+
> | Estima                      |                                         |
> | P.O. Box 1818               |  Voice: (847) 864-8772                  |
> | Evanston, IL 60204-1818     |  Fax:   (847) 864-6221                  |
> | U.S.A                       |  BBS:   (847) 864-8816                  |
> | e-mail: estima@estima.com   |  CompuServe: 73140,2202                 |
> |-----------------------------------------------------------------------|
> | Web Site:  http://www.estima.com                                      |
> | RATS Internet Mailing List: New members can join by sending e-mail to |
> |  MAISER@EFS.MQ.EDU.AU with the message: SUBSCRIBE  RATS-L             |
> +-----------------------------------------------------------------------+
Dear Sir,

I'm sorry that I still do not know where I can download the copy of
PKUNZIP from the web site estima.com. Would you like to tell me the
exact position of the copy of PKUNZIp, please? Thank you.

Regards,

KY Wu

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

From: WOO KAI YIN <hkwoo@hkabc.net>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Inquiry about PPunit root test
Date: Thu, 05 Feb 1998 02:35:54 +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>
X-Mailer: Mozilla 3.0 (Win95; I) (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Dear Sir,

I am required to input the lags variable in the function of @PPunit. The
lags variable is the same as truncation lag for the Newey-West
correction? Thank you for your response.

Regards,

Kevin

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

From: "Estima" <estima@estima.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: inquiry about Disco
Date: Wed, 4 Feb 1998 12:49:27 -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) v1.30)

> Dear Sir,
> 
> I'm sorry that I still do not know where I can download the copy of
> PKUNZIP from the web site estima.com. Would you like to tell me the
> exact position of the copy of PKUNZIp, please? Thank you.
> 
> Regards,
> 
> KY Wu
> 

It's on the Procedures and Examples Page, under "PKZipped files".

Tom Maycock
Estima
 
--
+-----------------------------+-----------------------------------------+
| Estima                      |                                         |
| P.O. Box 1818               |  Voice: (847) 864-8772                  |
| Evanston, IL 60204-1818     |  Fax:   (847) 864-6221                  | 
| U.S.A                       |  BBS:   (847) 864-8816                  |
| e-mail: estima@estima.com   |  CompuServe: 73140,2202                 |
|-----------------------------------------------------------------------|
| Web Site:  http://www.estima.com                                      |
| RATS Internet Mailing List: New members can join by sending e-mail to |
|  MAISER@EFS.MQ.EDU.AU with the message: SUBSCRIBE  RATS-L             |
+-----------------------------------------------------------------------+

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

From: WOO KAI YIN <hkwoo@hkabc.net>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: inquiry about Disco
Date: Thu, 05 Feb 1998 03:06:57 +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>
X-Mailer: Mozilla 3.0 (Win95; I) (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Estima wrote:
> 
> > Dear Sir,
> >
> > I'm sorry that I still do not know where I can download the copy of
> > PKUNZIP from the web site estima.com. Would you like to tell me the
> > exact position of the copy of PKUNZIp, please? Thank you.
> >
> > Regards,
> >
> > KY Wu
> >
> 
> It's on the Procedures and Examples Page, under "PKZipped files".
> 
> Tom Maycock
> Estima
> 
> -
Dear Sir,

Thank you for your help. I can find files including PKZ204G.EXE,
PROCO194.ZIP, PROCO594.ZIP. However, no one has reponse and the Netscape
said it could not receive any file. Would you like to help me once more
to solve my problem?

Thank you!

KY Wu

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

From: "Christopher F Baum" <baum@bc.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: inquiry about Disco
Date: Wed, 04 Feb 1998 14:13:19 -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) [1.3.2.1, s/n P020-300786-009] (via Mercury MTS (Bindery) v1.30)
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Mr Woo,

Do you realize that each phase of your conversation with Estima is being
broadcast worldwide to over 200 recipients (and over 199 of them probably do
not appreciate it)? PLEASE address your enquiries about services provided by
Estima to Estima directly! This sort of activity reduces the usefulness of
the RATS-L list for us all (and is likely to attract nasty comments about
how anyone who has used a desktop computer for more than three days can be
unaware of .ZIP files).

Kit Baum
Boston College

--On Thu, Feb 5, 1998 3:06 +0800 "WOO KAI YIN" <hkwoo@hkabc.net> wrote: 

> Estima wrote:
>> 
>> > Dear Sir,
>> >
>> > I'm sorry that I still do not know where I can download the copy of
>> > PKUNZIP from the web site estima.com. Would you like to tell me the
>> > exact position of the copy of PKUNZIp, please? Thank you.
>> >
>> > Regards,
>> >
>> > KY Wu
>> >
>> 
>> It's on the Procedures and Examples Page, under "PKZipped files".
>> 
>> Tom Maycock
>> Estima
>> 
>> -
> Dear Sir,
> 
> Thank you for your help. I can find files including PKZ204G.EXE,
> PROCO194.ZIP, PROCO594.ZIP. However, no one has reponse and the Netscape
> said it could not receive any file. Would you like to help me once more
> to solve my problem?
> 
> Thank you!
> 
> KY Wu



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

From: "Bird, Geoff" <BirdG@rba.gov.au>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Test - do not open
Date: Fri, 6 Feb 1998 11:29:00 +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 (5.0.1458.49) (via Mercury MTS (Bindery) v1.30)
Content-Type: text/plain



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

From: Pierre.SLADDEN@DG2.cec.be
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: winrats to unix-rats
Date: Fri, 6 Feb 1998 09:37:02 +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
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mercury MTS (Bindery) v1.30

Hello,

I'm searching for a solution (interface? frontend?) to transmit (link? relay?) directly WinRats data from Windows NT4 to Rats on  digital/alpha-unix. Solution may be based on VB or Excel (or ?) or may be a true client-server.

Thanks,

Pierre Sladden
European Commission
DG2 - Brussels

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

From: monia <monia@mizar.csic.es>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Inquiry about PPunit root test
Date: Fri, 06 Feb 1998 11:40:50 +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: Windows Eudora Light Version 3.0.1 (32) (via Mercury MTS (Bindery) v1.30)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"

At 02.35 5/2/98 +0800, you wrote:
>Dear Sir,
>
>I am required to input the lags variable in the function of @PPunit. The
>lags variable is the same as truncation lag for the Newey-West
>correction? Thank you for your response.
>
>Regards,
>
>Kevin

Yes is the same because in the PP test you dont need ti introduce a lag
variable.
Monia Ben Kaabia
Unidad de Economia Araria
Servicio de Investigacion Agroalimentaria
Diputacion General de Aragon
Apdo. 727
E-50080-Zaragoza (Spain)
Tel: ++ 34-976-576361
Fax: ++ 34-976-575501
monia@mizar.csic.es

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

From: John Cline <jcline@glinx.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: remove
Date: Fri, 6 Feb 1998 11:49:08 -0400
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) v1.30


Please remove me from your list

Amaranth And Associates &
Amaranth Co-operative Enterprises Ltd.
PO Box 448
Wolfville, Nova Scotia,  B0P 1X0
1-902-542-4002 ( Fax & Voice )
Email:  jcline@glinx.com
Homepage:  http://www.glinx.com/users/jcline/
Worker Owned

Environmental Products, Education, Sustainable Development And Consulting
For A Conserver Society



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

From: "Yong U. Glasure" <yglasure@wbu1.wbu.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Rolling Regression
Date: Fri, 6 Feb 1998 11:56:50 -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: Windows Eudora Light Version 1.5.2 (via Mercury MTS (Bindery) v1.30)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"

Dear Colleagues,

Can someone provide me a sample program of the rolling regression technique
that computes F-values? 

Thank you.

Yong U. Glasure
Associate Professor of Business Administration
Division of Business
Wayland Baptist University
Lubbock, Texas 79424
Voice:  806-794-8008
Fax:  806-794-4227
E-mail:  yglasure@wbu1.wbu.edu


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

From: "Estima" <estima@estima.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Rolling Regression
Date: Fri, 6 Feb 1998 12:46:04 -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) v1.30)

> 
> Can someone provide me a sample program of the rolling regression technique
> that computes F-values? 
> 

I'm not exactly sure what you are looking to with respect to 
computing F-values, but there are a couple of examples of doing 
rolling regressions on the RATS FAQ section of our web page 
(www.estima.com).

Sincerely,
Tom Maycock
Estima
 

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

From: Mark Thoma <mthoma@OREGON.UOREGON.EDU>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Rolling Regression
Date: Fri, 06 Feb 1998 12:04:54 -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>
MIME-version: 1.0
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32) (via Mercury MTS (Bindery) v1.30)
Content-type: text/plain; charset="us-ascii"

Try something like:

do i = 1965:1, 1997:12
   begin
   linreg(noprint) y 1960:1 i
   # constant x{1 to 4}
   exclude(noprint)
   # x{1 to 4}
   set xfstat i i = %cdstat
   end
end do i
*
*
graph(key=upleft) 1
# xfstat 65:1 97:12

I wrote the code quickly, so it may not be perfect, but hopefully you get
the idea.

Mark Thoma
Head, Economics
University of Oregon



At 12:46 PM 2/6/1998 -0600, you wrote:
>> 
>> Can someone provide me a sample program of the rolling regression technique
>> that computes F-values? 
>> 
>
>I'm not exactly sure what you are looking to with respect to 
>computing F-values, but there are a couple of examples of doing 
>rolling regressions on the RATS FAQ section of our web page 
>(www.estima.com).
>
>Sincerely,
>Tom Maycock
>Estima
> 
>

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

From: "Yong U. Glasure" <yglasure@wbu1.wbu.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Rolling Regression
Date: Fri, 6 Feb 1998 14:53:41 -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: Windows Eudora Light Version 1.5.2 (via Mercury MTS (Bindery) v1.30)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"

Dear Colleagues,

Earlier I asked for a sample of the rolling regression technique.  Professor
Toma of University of Oregon provided a sample of it.  Hence, please discard
my request.

Thank you for your assistance.



Yong U. Glasure
Associate Professor of Business Administration
Division of Business
Wayland Baptist University
Lubbock, Texas 79424
Voice:  806-794-8008
Fax:  806-794-4227
E-mail:  yglasure@wbu1.wbu.edu


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

From: Rob Trevor <robt@efs.mq.edu.au>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: winrats to unix-rats
Date: Sat, 7 Feb 1998 10:19:56 +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
Content-Type: text/plain; charset="us-ascii"
X-Mailer: Mercury MTS (Bindery) v1.30

At 7:37 PM +1100 6/2/98, Pierre.SLADDEN@DG2.cec.be wrote:
>...
>I'm searching for a solution (interface? frontend?) to transmit (link?
>relay?) directly WinRats data from Windows NT4 to Rats on
>digital/alpha-unix. Solution may be based on VB or Excel (or ?) or may be
>a true client-server.
>...

A simple solution might be a simple NSF server/client set-up. Since the v4
data files are binary compatible across all these platforms, you could
simply make NT file make the file available to the Unix box by NSF. (Or you
could script an FTP updating process, but that's a lot messier.)

Rob



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

From: Kenneth Leong <kleong@tartarus.uwa.edu.au>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Procedure for the Canova-Hansen (1995) Test.
Date: Mon, 9 Feb 1998 21:53:30 +0800 (WST)
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) v1.30


Hi,

Has anyone written a procedure to calculate the Canova-Hansen (1995,
JBES) test for seasonal unit roots?

 


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

From: jtebeka@oddo.fr
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: forcasting cointegrated system
Date: Mon, 9 Feb 1998 15:21:01 +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
Content-type: text/plain; charset=us-ascii
X-Mailer: Mercury MTS (Bindery) v1.30




I have estimated a cointegrated system using CATS, how can i proceed to
forecast it and to have basic statistics on forecast error (using THEIL
instruction).
Thanks.



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

From: "MR J G OTERO" <ecrdv@frost.csv.warwick.ac.uk>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Procedure for FM-OLS
Date: Mon, 9 Feb 1998 14:25:21 +0100 (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>
Organization: University of Warwick
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
X-mailer: Pegasus Mail for Windows (v2.23) (via Mercury MTS (Bindery) v1.30)

Hi,

Has anyone written a procedure to calculate the fully modified 
least-squares estimator of Phillips and Hansen (1990)?

Regards

Jesus G. Otero


****************************************
Jesus G. Otero
Department of Economics
University of Warwick
Coventry CV4 7AL
United Kingdom

e-mail: J.G.Otero@Warwick.ac.uk
Fax: 44 1203 523032
Tel: 44 1203 528240 (Office)
 

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

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

From: "B. Dan Wood" <bdanwood@polisci.tamu.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: RE: Procedure for FM-OLS
Date: Mon, 9 Feb 1998 09:21:58 -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
X-Mailer: Internet Mail Service (5.0.1457.3) (via Mercury MTS (Bindery) v1.30)
Content-Type: text/plain

Gauss has such a procedure in its coint module.

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

B. Dan Wood                                                Office: (409)
845-1610
Department of Political Science                   Home: (409) 690-0390
Texas A&M University                                  FAX:   (409)
847-8924
College Station, TX  77843-4348                 

HTTP://people.tamu.edu/~danwood/

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@


	-----Original Message-----
	From:	MR J G OTERO [SMTP:ecrdv@frost.csv.warwick.ac.uk]
	Sent:	Monday, February 09, 1998 7:25 AM
	To:	RATS Discussion List
	Subject:	Procedure for FM-OLS

	Hi,

	Has anyone written a procedure to calculate the fully modified 
	least-squares estimator of Phillips and Hansen (1990)?

	Regards

	Jesus G. Otero


	****************************************
	Jesus G. Otero
	Department of Economics
	University of Warwick
	Coventry CV4 7AL
	United Kingdom

	e-mail: J.G.Otero@Warwick.ac.uk
	Fax: 44 1203 523032
	Tel: 44 1203 528240 (Office)
	 

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

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

From: "Wai Lee"<lee_wai@jpmorgan.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Time-series and Panel regression
Date: Tue, 10 Feb 1998 10:34:36 -0400
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) v1.30


Dear colleagues:

Can I do both time-series and cross-sectional regression in the same
program in RATS?

Here is what I need to do.  Generate explanatory variable by time-series
rolling regression for each of the 20 y series for each month, say.  Then
for each month, I need to run a cross-sectional regression based on these
20 y variables and their generated-explanatory variables for that
particular month.  The parameters estimated for each month will then be
saved as time series for later analysis (Yes, you guess correctly.  This is
the Fama-MacBeth type of work).

Of course, I can do matrix computation for the cross-sectional regressions
for each month.  But I would like to know if there is any other easier way
to save my time.

Thanks for your help.

W. Lee




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

From: "N.A.Molinari" <nmolina@econ.wayne.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Time-series and Panel regression
Date: Tue, 10 Feb 1998 12:26:06 -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: Mozilla 4.01 [en] (Win95; I) (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

If you are combining cross-section with timeseries, then this is panel
data. Is there any reason you can't use :
CALENDAR(PANELOBS=5)
ALLOCATE 0 40//5
OPEN DATA PANEL.WKS
DATA(FORMAT=WKS,ORG=OBS) / Y Z1 Z2 X1 X2
*
And then do your analysis? This will allow you to take advantage of
timeseries as well as cross-sectional properties of the data when
estimating the parameters.
Wai Lee wrote:

> Dear colleagues:
>
> Can I do both time-series and cross-sectional regression in the same
> program in RATS?
>
> Here is what I need to do.  Generate explanatory variable by
> time-series
> rolling regression for each of the 20 y series for each month, say.
> Then
> for each month, I need to run a cross-sectional regression based on
> these
> 20 y variables and their generated-explanatory variables for that
> particular month.  The parameters estimated for each month will then
> be
> saved as time series for later analysis (Yes, you guess correctly.
> This is
> the Fama-MacBeth type of work).
>
> Of course, I can do matrix computation for the cross-sectional
> regressions
> for each month.  But I would like to know if there is any other easier
> way
> to save my time.
>
> Thanks for your help.
>
> W. Lee




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

From: Raphael Caillet <Raphael.Caillet@wanadoo.fr>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: garch.src
Date: Wed, 11 Feb 1998 09:48:13 +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 Light Version 3.0.3 (32) (via Mercury MTS (Bindery) v1.30)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"

hello,
I've a problem with the procedure : garch.src (last version) from the web
site estima.com 
when I use the instruction 
	source(noecho) garch.src
I've received the following message "error in the application" and I must
close Rats
I'm use Rats 3.4 on windows 3.11 with 32 Mo

Thank you for your assistance
--------------------------------------------------------
Raphael CAILLET
Chercheur doctorant
GAMMAP (Groupe d'Analyse des Marches de Matieres Premieres)
Universite Pierre Mendes-France
Espace Europe
BP 47 
38040 Grenoble Cedex 9 France
Tel 33-(0)476825841
Fax 33-(0)476825862
E-mail : gammap@upmf-grenoble.fr 

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

From: Katja Gerling <kgerling@ifw.uni-kiel.de>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: cross-equation restrictions in rats
Date: Wed, 11 Feb 1998 11:09:38 -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: The Kiel Institute of World Economics
X-Mailer: Mozilla 2.01 (Win16; I) (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

I have a question concerning estimating a system of 3 equations (with 
SUR): I would like to put some cross-equation-restrictions and in one of 
these, I want to set a coefficient to be equal in 2 of the 3 equations. 
Is this possible and if yes, how can I do that? The command 'equate' 
seems to work only if I set coefficients to be equal in all 3 
equations...
Katja

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

From: Katja Gerling <kgerling@ifw.uni-kiel.de>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: error message in system estimation
Date: Wed, 11 Feb 1998 12:52:28 -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: The Kiel Institute of World Economics
X-Mailer: Mozilla 2.01 (Win16; I) (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

while estimating a system of 3 equations with SUR and instrumental 
variables, I encountered the following error message: ## REG10. first 20 
instruments are linearly dependent over regression range.
nonetheless, rats did the estimation and the resulting parameter values 
seem reasonable. can somebody give me an explanation?
katja

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

From: "Wai Lee"<lee_wai@jpmorgan.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Time-series and Panel regression
Date: Wed, 11 Feb 1998 08:48:21 -0400
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) v1.30


Yes, but I still need 2 separate programs for the job, 1 for generating the
explanatory variables by time-series rolling regression; then another one
to read in the data again as panel data and do cross-sectional regressions.
In fact, I need to iterate the above procedures a few time to reduce small
sample bias, which makes it even more troublesome.

In fact, I have already got the job done by doing matrix algebra for the
cross-sectional regressions within the DO-loop of time-series rolling
regressions.  I am just wondering if there is an easier way for this that
we can share with other colleagues.  For e.g., SAS handles this with the
"by" command in "model."

Thanks for your suggestions anyway.

W. Lee





nmolina @ econ.wayne.edu ("N.A.Molinari") on 02/10/98 12:26:06 PM

To:   RATS-L @ efs.mq.edu.au ("RATS Discussion List") @ SMTP
cc:
Subject:  Re: Time-series and Panel regression



If you are combining cross-section with timeseries, then this is panel
data. Is there any reason you can't use :
CALENDAR(PANELOBS=5)
ALLOCATE 0 40//5
OPEN DATA PANEL.WKS
DATA(FORMAT=WKS,ORG=OBS) / Y Z1 Z2 X1 X2
*
And then do your analysis? This will allow you to take advantage of
timeseries as well as cross-sectional properties of the data when
estimating the parameters.
Wai Lee wrote:
> Dear colleagues:
>
> Can I do both time-series and cross-sectional regression in the same
> program in RATS?
>
> Here is what I need to do.  Generate explanatory variable by
> time-series
> rolling regression for each of the 20 y series for each month, say.
> Then
> for each month, I need to run a cross-sectional regression based on
> these
> 20 y variables and their generated-explanatory variables for that
> particular month.  The parameters estimated for each month will then
> be
> saved as time series for later analysis (Yes, you guess correctly.
> This is
> the Fama-MacBeth type of work).
>
> Of course, I can do matrix computation for the cross-sectional
> regressions
> for each month.  But I would like to know if there is any other easier
> way
> to save my time.
>
> Thanks for your help.
>
> W. Lee









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

From: "Christopher F Baum" <baum@bc.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: cross-equation restrictions in rats
Date: Wed, 11 Feb 1998 10:07:41 -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) [1.3.2.2, s/n P020-300786-009] (via Mercury MTS (Bindery) v1.30)
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

RESTRICT and MRESTRICT may both be used for this purpose. Since the SUR
coefficients are a single long beta-vector, they may be referred to (as
indeed they are in the SUR output) by number, and restrictions imposed
accordingly.

Kit Baum
 
--On Wed, Feb 11, 1998 11:09 -0800 "Katja Gerling"
<kgerling@ifw.uni-kiel.de> wrote: 

> I have a question concerning estimating a system of 3 equations (with 
> SUR): I would like to put some cross-equation-restrictions and in one of 
> these, I want to set a coefficient to be equal in 2 of the 3 equations. 
> Is this possible and if yes, how can I do that? The command 'equate' 
> seems to work only if I set coefficients to be equal in all 3 
> equations...
> Katja



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

From: "Estima" <estima@estima.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: cross-equation restrictions in rats
Date: Wed, 11 Feb 1998 09:56:54 -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) v1.30)


> RESTRICT and MRESTRICT may both be used for this purpose. Since the SUR
> coefficients are a single long beta-vector, they may be referred to (as
> indeed they are in the SUR output) by number, and restrictions imposed
> accordingly.
> 

There's an example of this on page 5-26 of the manual. This 
particular version iterated SUR, so is more complicated than if you 
are just doing a single estimation, but should be helpful.

The basic sequence is:

1. Do SUR (unrestricted)
2. Use RESTRICT(REPLACE) or MRESTRICT(REPLACE) to impose the 
restriction
3. Do SUR(CREATE) to generate output displaying the results of the 
restricted estimation.

Tom Maycock
Estima

--
+-----------------------------+-----------------------------------------+
| Estima                      |                                         |
| P.O. Box 1818               |  Voice: (847) 864-8772                  |
| Evanston, IL 60204-1818     |  Fax:   (847) 864-6221                  | 
| U.S.A                       |  BBS:   (847) 864-8816                  |
| e-mail: estima@estima.com   |  CompuServe: 73140,2202                 |
|-----------------------------------------------------------------------|
| Web Site:  http://www.estima.com                                      |
| RATS Internet Mailing List: New members can join by sending e-mail to |
|  MAISER@EFS.MQ.EDU.AU with the message: SUBSCRIBE  RATS-L             |
+-----------------------------------------------------------------------+

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

From: "Estima" <estima@estima.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Time-series and Panel regression
Date: Wed, 11 Feb 1998 09:56:53 -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) v1.30)

> 
> Yes, but I still need 2 separate programs for the job, 1 for generating the
> explanatory variables by time-series rolling regression; then another one
> to read in the data again as panel data and do cross-sectional regressions.
> In fact, I need to iterate the above procedures a few time to reduce small
> sample bias, which makes it even more troublesome.
> 
> In fact, I have already got the job done by doing matrix algebra for the
> cross-sectional regressions within the DO-loop of time-series rolling
> regressions.  I am just wondering if there is an easier way for this that
> we can share with other colleagues.  For e.g., SAS handles this with the
> "by" command in "model."
> 
 
If I understand what you want to do, this can be easily handled using 
a dummy variable and the SMPL option. For example, suppose you have 
panel data with 10 individuals and 50 observations per individual, so 
that your CAL/ALL looks like:

cal(panel=50)
all 10//50

If you want to do a regression on the entire panel, you would do:

linreg y
# x1 x2

If you want to do a regression on only one individual, you would do 
something like:

linreg y 1//1 1//50
# x1 x2

You could loop over individuals by doing:

do indiv=1,10
 linreg y indiv//1 indiv//50
 # x1 x2
 * bookkeeping instructions here
end do

To do a regression using data from all individuals for a 
particular time period, do something like:

set indivdummy = %indiv(t)==5
linreg(smpl=indivdummy) y
# x1 x2

In a loop, this would be:

do indiv=1,10
 set indivdummy = %indiv(t)==indiv
 linreg(smpl=indivdummy) y
 # x1 x2
end do 

Sincerely,
Tom Maycock
--
+-----------------------------+-----------------------------------------+
| Estima                      |                                         |
| P.O. Box 1818               |  Voice: (847) 864-8772                  |
| Evanston, IL 60204-1818     |  Fax:   (847) 864-6221                  | 
| U.S.A                       |  BBS:   (847) 864-8816                  |
| e-mail: estima@estima.com   |  CompuServe: 73140,2202                 |
|-----------------------------------------------------------------------|
| Web Site:  http://www.estima.com                                      |
| RATS Internet Mailing List: New members can join by sending e-mail to |
|  MAISER@EFS.MQ.EDU.AU with the message: SUBSCRIBE  RATS-L             |
+-----------------------------------------------------------------------+

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

From: Anatoly Karp <karp@math.wisc.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: [Q]: Simple (?) question on ARCH-M 
Date: Wed, 11 Feb 1998 13:10:11 -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: Mercury MTS (Bindery) v1.30

Hello everybody,

My question probably has a simple answer, but I've been
unable to solve it:

Let's say we're estimating a simple ARCH-M model

y(t) = a + b*h(t) + u(t);     var(u(t)) = h(t)
h(t) = c + d*h(t-1) + e(t)

As far as I understand, we estimate the parameter
vector (a,b,c,d) using, say, MLE,  and only the
observations of y (indexed 1...T) are supplied on 
input. Now what? - 

Suppose I want to forecast y(T+1) or h(T+1).
It seems I'm stuck since in either case I need
(an estimate of) h(T), then h(T-1), h(T-2) ...
etc. 

So my question is: how could this ever be implemented?

Any help is greatly appreciated.

Thanks, 
Anatoly 



-- 
Anatoly Karp                        e-mail: karp@math.wisc.edu
Department of Mathematics           URL:    http://www.math.wisc.edu/~karp
UW-Madison                          phone:  (608) 258-1786

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

From: "Wai Lee"<lee_wai@jpmorgan.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: [Q]: Simple (?) question on ARCH-M
Date: Thu, 12 Feb 1998 09:17:37 -0400
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) v1.30


When you fit the model, you get the fitted time series for h(t), h(t-1),
........  Then use h(t) to forecast h(t+1) by your GARCH equation.  Finally,
use the forecasted h(t+1) to forecast y(t+1) in your GARCH-in-mean
equation.

I hope I dont misinterpret your question.





karp @ math.wisc.edu (Anatoly Karp) on 02/11/98 01:10:11 PM

To:   RATS-L @ efs.mq.edu.au ("RATS Discussion List") @ SMTP
cc:
Subject:  [Q]: Simple (?) question on ARCH-M



Hello everybody,
My question probably has a simple answer, but I've been
unable to solve it:
Let's say we're estimating a simple ARCH-M model
y(t) = a + b*h(t) + u(t);     var(u(t)) = h(t)
h(t) = c + d*h(t-1) + e(t)
As far as I understand, we estimate the parameter
vector (a,b,c,d) using, say, MLE,  and only the
observations of y (indexed 1...T) are supplied on
input. Now what? -
Suppose I want to forecast y(T+1) or h(T+1).
It seems I'm stuck since in either case I need
(an estimate of) h(T), then h(T-1), h(T-2) ...
etc.
So my question is: how could this ever be implemented?
Any help is greatly appreciated.
Thanks,
Anatoly


--
Anatoly Karp                        e-mail: karp@math.wisc.edu
Department of Mathematics           URL:    http://www.math.wisc.edu/~karp
UW-Madison                          phone:  (608) 258-1786







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

From: Klaus Fischer <Klaus.Fischer@fas.ulaval.ca>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Time-series and Panel regression
Date: Tue, 03 Feb 1998 14:47:28 -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: Mozilla 4.01 [en] (Win95; I) (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="------------AB666D74951776F0A2222BF9"

This is a multi-part message in MIME format.
--------------AB666D74951776F0A2222BF9
Content-Type: text/plain; charset=iso-8859-1
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit

If I understand your problem correctly, Martin Desrochers, a masters
student who is battling through the same problem, found the "solution"
to the problem in page 5.3 of the Rats Manual (Subsample within a
panel). The problem is that he is trying it out and in his program IT
DOESN´T WORK, and we are doing some maximum likelihood estimations so
your "matrix solution" gets complicated.. We are about to contact Tom
with our case. Let you know whatever comes out of this.

Klaus Fischer
Laval University

Wai Lee wrote:

> Yes, but I still need 2 separate programs for the job, 1 for
> generating the
> explanatory variables by time-series rolling regression; then another
> one
> to read in the data again as panel data and do cross-sectional
> regressions.
> In fact, I need to iterate the above procedures a few time to reduce
> small
> sample bias, which makes it even more troublesome.
>
> In fact, I have already got the job done by doing matrix algebra for
> the
> cross-sectional regressions within the DO-loop of time-series rolling
> regressions.  I am just wondering if there is an easier way for this
> that
> we can share with other colleagues.  For e.g., SAS handles this with
> the
> "by" command in "model."
>
> Thanks for your suggestions anyway.
>
> W. Lee
>
> nmolina @ econ.wayne.edu ("N.A.Molinari") on 02/10/98 12:26:06 PM
>
> To:   RATS-L @ efs.mq.edu.au ("RATS Discussion List") @ SMTP
> cc:
> Subject:  Re: Time-series and Panel regression
>
> If you are combining cross-section with timeseries, then this is panel
>
> data. Is there any reason you can't use :
> CALENDAR(PANELOBS=5)
> ALLOCATE 0 40//5
> OPEN DATA PANEL.WKS
> DATA(FORMAT=WKS,ORG=OBS) / Y Z1 Z2 X1 X2
> *
> And then do your analysis? This will allow you to take advantage of
> timeseries as well as cross-sectional properties of the data when
> estimating the parameters.
> Wai Lee wrote:
> > Dear colleagues:
> >
> > Can I do both time-series and cross-sectional regression in the same
>
> > program in RATS?
> >
> > Here is what I need to do.  Generate explanatory variable by
> > time-series
> > rolling regression for each of the 20 y series for each month, say.
> > Then
> > for each month, I need to run a cross-sectional regression based on
> > these
> > 20 y variables and their generated-explanatory variables for that
> > particular month.  The parameters estimated for each month will then
>
> > be
> > saved as time series for later analysis (Yes, you guess correctly.
> > This is
> > the Fama-MacBeth type of work).
> >
> > Of course, I can do matrix computation for the cross-sectional
> > regressions
> > for each month.  But I would like to know if there is any other
> easier
> > way
> > to save my time.
> >
> > Thanks for your help.
> >
> > W. Lee



--------------AB666D74951776F0A2222BF9
Content-Type: text/x-vcard; charset=us-ascii; name="vcard.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Klaus P. Fischer
Content-Disposition: attachment; filename="vcard.vcf"

begin:          vcard
fn:             Klaus P. Fischer
n:              Fischer;Klaus P.
org:            Centre de recherche en économie et finance appliquée 
email;internet: Klaus.Fischer@fas.ulaval.ca
title:          Université Laval, Quebec, CANADA
note:           on sabbatical at: FOGAFIN, Carrera 7, No 35-40, Santa Fe de Bogotá, COLOMBIA
x-mozilla-cpt:  ;0
x-mozilla-html: FALSE
end:            vcard


--------------AB666D74951776F0A2222BF9--


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

From: Alfonso.ARPAIA@EUROSTAT.cec.be
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Cointegration
Date: Thu, 12 Feb 1998 17:36:27 +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
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mercury MTS (Bindery) v1.30

Hello everybody,
I'm making a cointegration analysis in cats. I have a simple question 
Before testing for reduced rank do I have to test the order of integration 
(e.g. using DF/ADF,PP test) of each variable that enters the VAR or it is sufficient 
to check if the var is stable looking at the eigenvalues of the companion matrix? 

many thanks 
Alfonso

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

From: Kenneth Leong <kleong@tartarus.uwa.edu.au>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Lucas (1980) Filter.
Date: Fri, 13 Feb 1998 15:11:29 +0800 (WST)
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) v1.30


Hi all,

In a 1980 AER paper, Lucas proposed a filter which extracts a long-run
signal from a time series. The filter is:

X(t)(A) = [(1-A)/(1+A)].Sum(from k=-n to n)(A^|k|).X(t+k) 

where X(t) is the variable of interest, X(t)(A) is the filtered series, 
A is a parameter restricted to be between 0 and 1, n is the sample
size and k is the index of the summation.

In words, the filtered series is the factor (1-A)/(1+A) times a summation
(from -n to n) of the parameter A raised to the power of the absolute
value of (k) times X(t+k).

As A approaches 0, no filtering occurs.
As A approaches 1, the filtered X(t)(A) approach the sample mean of the
original series.

I am just wondering how I could create this filter in RATS. Do I need some
sort of loop?

Many thanks.

Regards,

Kenneth.


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

From: Luca Cazzulani <luca.cazzulani@uni-bocconi.it>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Lucas (1980) Filter.
Date: Mon, 16 Feb 1998 17:56:52 +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: Universita' Bocconi - Milano (Italy)
X-Mailer: Windows Eudora Pro Version 2.2 (16) (via Mercury MTS (Bindery) v1.30)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"

At 15.11 13/02/98 +0800, you wrote:
>
>Hi all,
>
>In a 1980 AER paper, Lucas proposed a filter which extracts a long-run
>signal from a time series. The filter is:
>
>X(t)(A) = [(1-A)/(1+A)].Sum(from k=-n to n)(A^|k|).X(t+k) 
>
>where X(t) is the variable of interest, X(t)(A) is the filtered series, 
>A is a parameter restricted to be between 0 and 1, n is the sample
>size and k is the index of the summation.
>
>In words, the filtered series is the factor (1-A)/(1+A) times a summation
>(from -n to n) of the parameter A raised to the power of the absolute
>value of (k) times X(t+k).
>
>As A approaches 0, no filtering occurs.
>As A approaches 1, the filtered X(t)(A) approach the sample mean of the
>original series.
>
>I am just wondering how I could create this filter in RATS. Do I need some
>sort of loop?
>
>Many thanks.
>
>Regards,
>
>Kenneth.
>
>
Could you give me a detailed bibliography of the article containing the
filter you talk about in your message?

Many thanks in advance

Luca Cazzulani


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

From: Kenneth Leong <kleong@tartarus.uwa.edu.au>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Lucas (1980) Filter.
Date: Tue, 17 Feb 1998 10:03:49 +0800 (WST)
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) v1.30


> >In a 1980 AER paper, Lucas proposed a filter which extracts a long-run
> >signal from a time series. The filter is:
> >
> >X(t)(A) = [(1-A)/(1+A)].Sum(from k=-n to n)(A^|k|).X(t+k) 
> >
> >where X(t) is the variable of interest, X(t)(A) is the filtered series, 
> >A is a parameter restricted to be between 0 and 1, n is the sample
> >size and k is the index of the summation.
> >
> >As A approaches 0, no filtering occurs.
> >As A approaches 1, the filtered X(t)(A) approach the sample mean of the
> >original series.
> >
> >I am just wondering how I could create this filter in RATS. Do I need some
> >sort of loop?

--------------------------------
> Could you give me a detailed bibliography of the article containing the
> filter you talk about in your message?
--------------------------------
Hi Luca,

One reference is James Bullard's 'Measures of Money and the
Quantity Theory' in the Jan/Feb 1994 Federal Reserve Bank of St. Louis
Review, pp. 19-30.

The filter was originally proposed by Lucas in 'Two Illustrations of the
Quantity Theory of Money', AER, December 1980, pp. 1005-14.

I was wondering if the Hodrick-Prescott filter can be used instead of 
the Lucas filter. The Lucas filter extracts the long-run signal from a
time series, while the HP filter extracts the trend.

Thanks.

Kenneth.


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

From: jtebeka@oddo.fr
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Daily data
Date: Tue, 17 Feb 1998 10:10:39 +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
Content-type: text/plain; charset=us-ascii
X-Mailer: Mercury MTS (Bindery) v1.30




I have daily data. I read my data file without any problem but when i am
trying to use GARCH.SRC to estimate a garch model the program stops at the
first #na date. How can i proceed to estimate Garch model with #na data ?
Jacques



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

From: "Wai Lee"<lee_wai@jpmorgan.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Daily data
Date: Tue, 17 Feb 1998 08:45:03 -0400
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) v1.30







jtebeka @ oddo.fr on 02/17/98 10:10:39 AM

To:   RATS-L @ efs.mq.edu.au ("RATS Discussion List") @ SMTP
cc:
Subject:  Daily data


I write my own code for GARCH estimations and thus don't know much about
the GARCH.SRC.  Is it possible that you ran into the same problem discussed
on p5-31 of the manual?  Did you reserve a few (depending on your lags in
GARCH) observations in your estimation?



I have daily data. I read my data file without any problem but when i am
trying to use GARCH.SRC to estimate a garch model the program stops at the
first #na date. How can i proceed to estimate Garch model with #na data ?
Jacques








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

From: jtebeka@oddo.fr
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: =?iso-8859-1?Q?R=E9f._:_Re:_Daily_data?=
Date: Tue, 17 Feb 1998 14:58:02 +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
Content-type: text/plain; charset=iso-8859-1
Content-transfer-encoding: quoted-printable
X-Mailer: Mercury MTS (Bindery) v1.30




My #na data correspond to hollyday. They are not at the begining of my
sample.




lee_wai@jpmorgan.com sur 17/02/98 13:45:03

Veuillez r=E9pondre =E0 RATS-L@efs.mq.edu.au

Pour :    RATS-L@efs.mq.edu.au
cc :   (ccc : Jacques Tebeka/ODDO)
Objet :   Re: Daily data


=

Content-type: text/plain; charset=us-ascii








jtebeka @ oddo.fr on 02/17/98 10:10:39 AM
To:   RATS-L @ efs.mq.edu.au ("RATS Discussion List") @ SMTP
cc:
Subject:  Daily data

I write my own code for GARCH estimations and thus don't know much about
the GARCH.SRC.  Is it possible that you ran into the same problem discussed
on p5-31 of the manual?  Did you reserve a few (depending on your lags in
GARCH) observations in your estimation?


I have daily data. I read my data file without any problem but when i am
trying to use GARCH.SRC to estimate a garch model the program stops at the
first #na date. How can i proceed to estimate Garch model with #na data ?
Jacques














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

From: David Watt <dwatt@bank-banque-canada.ca>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: =?iso-8859-1?Q?R=E9f=2E?= : Re: Daily data
Date: Tue, 17 Feb 1998 09:12:44 -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: Mozilla 4.04 [en] (X11; I; SunOS 5.5.1 sun4u) (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

The best way to estimate the GARCH model for daily
data is to collapse the series to remove the missing
observations. This is best done using the SAMPLE
command and the example on 14-209 of the Manual

The recursive nature of the GARCH model does not
handle missing observations. You have to collapse the
series in a preliminary step.

Dave.


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

From: jtebeka@oddo.fr
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: =?iso-8859-1?Q?R=E9f._:_Re:_R=E9f._:_Re:_Daily_data?=
Date: Tue, 17 Feb 1998 16:46:23 +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
Content-type: text/plain; charset=iso-8859-1
Content-transfer-encoding: quoted-printable
X-Mailer: Mercury MTS (Bindery) v1.30




It seems to work but imagine that i make some forecast how can i have a=

control on the true date ?




dwatt@bank-banque-canada.ca sur 17/02/98 15:12:44

Veuillez r=E9pondre =E0 RATS-L@efs.mq.edu.au

Pour :    RATS-L@efs.mq.edu.au
cc :   (ccc : Jacques Tebeka/ODDO)
Objet :   Re: R=E9f. : Re: Daily data


=

Content-type: text/plain; charset=us-ascii


The best way to estimate the GARCH model for daily
data is to collapse the series to remove the missing
observations. This is best done using the SAMPLE
command and the example on 14-209 of the Manual
The recursive nature of the GARCH model does not
handle missing observations. You have to collapse the
series in a preliminary step.
Dave.








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

From: Luca Cazzulani <luca.cazzulani@uni-bocconi.it>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Rolling Regression
Date: Wed, 18 Feb 1998 16:23:12 +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: Universita' Bocconi - Milano (Italy)
X-Mailer: Windows Eudora Pro Version 2.2 (16) (via Mercury MTS (Bindery) v1.30)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"

I am estimating a regression over a large sample , using a rolling technique.
I think it's likely that for some of the samples in which the regression is
estimated one or more of the regressors will be not significantly different
from zero. Is there a way to tell rats to carry on the estimation using each
time only the significant regressores?

Many thanks in advance

Luca Cazzulani


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

From: WOO KAI YIN <hkwoo@hkabc.net>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: INQUIRY ABOUT ERROR IN RATS PROCEDURE
Date: Sun, 22 Feb 1998 01:37:31 +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>
X-Mailer: Mozilla 3.0 (Win95; I) (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Dear Sir,

I've two problems. First of all, while I run a restriction test of the
subset of beta in CATS, an error message, floating point invalid,
SOMETIMES appears. May I ask what does such message mean? How can I
avoid it? Thank you.

Secondly, I am testing for cointegration of another model with a max.
number of dummies up to nine. I know that DisCo program may deal with it
.. I have the program but I find (in fact, I GUESS) the default max.
number of drift function allowed in the DisCo program comes to three
only. Anyone can let me know whether I misunderstood the manual of the
program? And How can I change the default number? I would be very much
grateful if you can give me assistance!

Thank you.

Regards,

Kevin

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

From: WOO KAI YIN <hkwoo@hkabc.net>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: inquiry about error messages in CATS 
Date: Sun, 22 Feb 1998 01:42:26 +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>
X-Mailer: Mozilla 3.0 (Win95; I) (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Dear Sir,

I've two problems. First of all, while I run a restriction test of the
subset of beta in CATS, an error message, floating point invalid,
SOMETIMES appears. May I ask what does such message mean? How can I
avoid it? Thank you.

Secondly, I am testing for cointegration of another model with a max.
number of dummies up to nine. I know that DisCo program may deal with it
.. I have the program but I find (in fact, I GUESS) the default max.
number of drift function allowed in the DisCo program comes to three
only. Anyone can let me know whether I misunderstood the manual of the
program? And How can I change the default number? I would be very much
grateful if you can give me assistance!

Thank you.

Regards,

Kevin

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

From: Katja Gerling <kgerling@ifw.uni-kiel.de>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: (no subject)
Date: Mon, 23 Feb 1998 11:59:56 -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: The Kiel Institute of World Economics
X-Mailer: Mozilla 2.01 (Win16; I) (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

i have a very simple question: how can i save the residuals of the 
equations in a system of equations estimated by sur?
katja

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

From: Jongchurl Kim <kim.317@osu.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: (no subject)
Date: Mon, 23 Feb 1998 06:55:57 -0500 (EST)
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: Windows Eudora Pro Version 2.1.2 (via Mercury MTS (Bindery) v1.30)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"

If the system has three equations, first you need to difine each equation in
lin reg instruction as follows:the system could take any forms depending on
the property of data series you are using.

linreg(define=1) var1
# constant var1{1 to p} var2{1 to p} var{1 to p}
linreg(define=2) var2
# constant var1{1 to p} var2{1 to p}
linreg(define=3) var3
# constant var1{1 to p} var3{1 to p}

Then, sur instruction needs supplementary cards to store the residuals from
equation 1 in series 1, equation 2 in series 2, and equation 3 in series 3.

sur 3 /
# 1 1
# 2 2
# 3 3

One more thing, when you call in your data, you need to reserve spaces for
resisual series, the number of spaces should be at least 2 times the number
of residual series.
example) 
  cal 1980:1
  all 6 1996:12 ; since there are three residual series, you need to reserve
6                   spaces.
  


At 11:59 AM 2/23/98 -0800, you wrote:
>i have a very simple question: how can i save the residuals of the 
>equations in a system of equations estimated by sur?
>katja
>
>

   =============================================================================
   Jongchurl Kim                                     Department of Economics
   kim.317@osu.edu                                   The Ohio State University
   Office (614)292-5765                              311 Arps Hall
          (614)292-3906(FAX)                         1945 North High Street
                                                     Columbus,Ohio 43210-1172


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

From: "Estima" <estima@estima.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: (no subject)
Date: Mon, 23 Feb 1998 09:35:52 -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) v1.30)

> Then, sur instruction needs supplementary cards to store the residuals from
> equation 1 in series 1, equation 2 in series 2, and equation 3 in series 3.
> 
> sur 3 /
> # 1 1
> # 2 2
> # 3 3

As Jongchurl notes here, you can certainly use series numbers to 
specify the series on the SUR supplementary cards. And, if you do so, 
you need to be careful to set aside the appropriate number of series 
so you don't overwrite actual data.

However, unless you really need to use series numbers (e.g., to
simplify further automated processing of the residuals), you may
find it easier to simply specify new series _names_ for the
residuals. For example:

SUR 3
# 1 resids1
# 2 resids2
# 3 resids3

Now you can work with these series by name. For example:

print / resids1 resids2 resids3

and so on. Also, this way you don't have to worry about declaring
the series ahead of time. 

Sincerely,
Tom Maycock
Estima
--
+-----------------------------+-----------------------------------------+
| Estima                      |                                         |
| P.O. Box 1818               |  Voice: (847) 864-8772                  |
| Evanston, IL 60204-1818     |  Fax:   (847) 864-6221                  | 
| U.S.A                       |  BBS:   (847) 864-8816                  |
| e-mail: estima@estima.com   |  CompuServe: 73140,2202                 |
|-----------------------------------------------------------------------|
| Web Site:  http://www.estima.com                                      |
| RATS Internet Mailing List: New members can join by sending e-mail to |
|  MAISER@EFS.MQ.EDU.AU with the message: SUBSCRIBE  RATS-L             |
+-----------------------------------------------------------------------+

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

From: "Wai Lee"<lee_wai@jpmorgan.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: ESMOOTH
Date: Mon, 23 Feb 1998 17:31:00 -0400
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) v1.30


I am not quite sure whether I understand the ESMOOTH command correctly or
not, and I don't have Gardner (1985) cited in the manual.

I need to create a new series y from existing variable x as

y(t) = x(t-1) + b*x(t-2) + b**2*x(t-3) + b**3*x(t-4) + ..............

where b = 0.5, say.
Should I just do

ESMOOTH(TREND=EXPONENTIAL,ALPHA=0.5,SMOOTHED=Y)    X

Thanks.





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

From: "Estima" <estima@estima.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: ESMOOTH
Date: Mon, 23 Feb 1998 17:02:51 -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) v1.30)

> I am not quite sure whether I understand the ESMOOTH command correctly or
> not, and I don't have Gardner (1985) cited in the manual.
> 
> I need to create a new series y from existing variable x as
> 
> y(t) = x(t-1) + b*x(t-2) + b**2*x(t-3) + b**3*x(t-4) + ..............
> 
> where b = 0.5, say.
> Should I just do
> 
> ESMOOTH(TREND=EXPONENTIAL,ALPHA=0.5,SMOOTHED=Y)    X
> 

Uh, no I don't think so. 

It sounds like you want to do this using either SET or FILTER
(whichever you find easier). This is assuming that the "+ ........"
above just means some higher number of lags, not an infinite number
of lags. 

Sincerely,
Tom Maycock
Estima
--
+-----------------------------+-----------------------------------------+
| Estima                      |                                         |
| P.O. Box 1818               |  Voice: (847) 864-8772                  |
| Evanston, IL 60204-1818     |  Fax:   (847) 864-6221                  | 
| U.S.A                       |  BBS:   (847) 864-8816                  |
| e-mail: estima@estima.com   |  CompuServe: 73140,2202                 |
|-----------------------------------------------------------------------|
| Web Site:  http://www.estima.com                                      |
| RATS Internet Mailing List: New members can join by sending e-mail to |
|  MAISER@EFS.MQ.EDU.AU with the message: SUBSCRIBE  RATS-L             |
+-----------------------------------------------------------------------+

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

From: jtebeka@oddo.fr
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Read quaterly data
Date: Wed, 25 Feb 1998 15:04:03 +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
Content-type: text/plain; charset=us-ascii
X-Mailer: Mercury MTS (Bindery) v1.30




I would like to read a quaterly data file that has this structure :
               x
Q1 70     22
Q2 70     21
How can i proceed. Rats can not recognize that date format and i do not
want to scratch the dates.
Jacques.



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

From: "Stuart M. Glosser" <glossers@uwwvax.uww.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Read quaterly data
Date: Wed, 25 Feb 1998 09:22:47 -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) v1.30

>I would like to read a quaterly data file that has this structure :
>               x
>Q1 70     22
>Q2 70     21
>How can i proceed. Rats can not recognize that date format and i do not
>want to scratch the dates.

Jacques,

One quick solution, prior to having Rats read the data, is to use a spread
sheet program, such as Excel, to open the initial file that contains the
dates and data (I assume it is a text file). Excel's "Text to Columns"
feature would enable you to put the dates and the actual data into
different columns. Then copy the column of actual data into a text file or
paste it directly into the Rats program that was going to read the data.

Stuart Glosser
Dept. of Economics
Univ. of Wisconsin at Whitewater
Whitewater, Wi. 53190




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