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 PAA101162
	for <baum@monet.bc.edu>; Thu, 3 Dec 1998 15:35:30 -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 PAA691236
	for baum@mail1.bc.edu; Thu, 3 Dec 1998 15:35:27 -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 PAA577210
	for <baum@bc.edu>; Thu, 3 Dec 1998 15:35:01 -0500
Received: from efs1.efs.mq.edu.au (efs1.efs.mq.edu.au [137.111.64.8])
	by sunb.ocs.mq.edu.au (8.9.0/8.9.0) with ESMTP id HAA13420
	for <baum@bc.edu>; Fri, 4 Dec 1998 07:34:53 +1100 (EST)
Received: from EFS1/SpoolDir by efs1.efs.mq.edu.au (Mercury 1.40);
    4 Dec 98 07:36:00 GMT+1000
Received: from SpoolDir by EFS1 (Mercury 1.40); 4 Dec 98 07:35:58 GMT+1000
To: baum@bc.edu
Date: Fri, 4 Dec 98 7:35:57 GMT+1000
Subject: Re: 
Message-ID: <27C2F27DEA@efs1.efs.mq.edu.au>

From: Hung-Jen Wang <hjwang@ieas.econ.sinica.edu.tw>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: a switch between OLS and 2SLS
Date: Mon, 02 Nov 1998 19:42:21 +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: QUALCOMM Windows Eudora Light Version 3.0.5 (32) (via Mercury MTS (Bindery) v1.40)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"

Hello,

     I'd like to ask a hopefully simple programming problem.  In my
program I wish to let users have an option of choosing between OLS
estimation and 2SLS (IV).  Is there a simple way to implement such a
"switch"? Ideally, the program should look like the following:

===========

<read in data and so on>

   ** Customization Section **
compute method = 1   ; * 1: OLS, 2: IV
.
.
   ** end of customization **
.
.
<skip main body>
.
.
   ** begin estimation **

if method ==2   ; * that is, you choose to do IV
  {
  compute [string] widget = 'inst'
  instruments z1 z2 z3 z4
  }
else
  {
  compute [string] widget = ' '
  }
end if


linreg(widget) y
# x1 x2 x3 x4

======================

It doesn't work because of the string variable "widget".  Rates
returns something like this:

     linreg(widget)<<<<
     ## OP3. This Instruction Does Not Have An Option wid
     >>>>

Any suggestion toward solving the problem will be appreciated!



--------------------------------------------------------------------------
Hung-Jen Wang                     internet: hjwang@ieas.econ.sinica.edu.tw
Institute of Economics            Phone: 886-2-26533595 ext 428             
Academia Sinica                   Fax:   886-2-27853946 
Taipei 115, Taiwan
--------------------------------------------------------------------------



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

From: "Christopher F Baum" <baum@bc.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: a switch between OLS and 2SLS
Date: Mon, 02 Nov 1998 09:10:18 -0500
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
X-Mailer: Mulberry (MacOS) [1.3.4, s/n P020-300786-009] (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Why not just include the linreg command within the if-then-else statement?
Or, if you have stuff to do in the meanwhile, just repeat the if, switching
on method?

Kit Baum

--On Mon, Nov 2, 1998 19:42 +0800 "Hung-Jen Wang"
<hjwang@ieas.econ.sinica.edu.tw> wrote: 

> Hello,
> 
>      I'd like to ask a hopefully simple programming problem.  In my
> program I wish to let users have an option of choosing between OLS
> estimation and 2SLS (IV).  Is there a simple way to implement such a
> "switch"? Ideally, the program should look like the following:
> 
> ===========
> 
> <read in data and so on>
> 
>    ** Customization Section **
> compute method = 1   ; * 1: OLS, 2: IV
> .
> .
>    ** end of customization **
> .
> .
> <skip main body>
> .
> .
>    ** begin estimation **
> 
> if method ==2   ; * that is, you choose to do IV
>   {
>   compute [string] widget = 'inst'
>   instruments z1 z2 z3 z4
>   }
> else
>   {
>   compute [string] widget = ' '
>   }
> end if
> 
> 
> linreg(widget) y
> # x1 x2 x3 x4
> 
> ======================
> 
> It doesn't work because of the string variable "widget".  Rates
> returns something like this:
> 
>      linreg(widget)<<<<
>      ## OP3. This Instruction Does Not Have An Option wid
>      >>>>
> 
> Any suggestion toward solving the problem will be appreciated!
> 
> 
> 
> --------------------------------------------------------------------------
> Hung-Jen Wang                     internet: hjwang@ieas.econ.sinica.edu.tw
> Institute of Economics            Phone: 886-2-26533595 ext 428           
 
> Academia Sinica                   Fax:   886-2-27853946 
> Taipei 115, Taiwan
> --------------------------------------------------------------------------
> 
> 



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

From: jtebeka@oddo.fr
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Kalman filter
Date: Mon, 2 Nov 1998 16:12:26 +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) v1.40




How can i estimate a model base on kalman filtering.
Thanks for your help.
Jacques.



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

From: "Estima" <estima@estima.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Kalman filter
Date: Mon, 2 Nov 1998 10:18:26 -0600
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
X-mailer: Pegasus Mail for Win32 (v2.54) (via Mercury MTS (Bindery) v1.40)

> 
> 
> How can i estimate a model base on kalman filtering.
> Thanks for your help.

Jacques:

There's an entire chapter on Kalman filtering in the RATS manual--you 
might want to start there (also see the descriptions of KALKMAN, 
KFSET, TVARYING in Chapter 14).

Sincerely,
Tom Maycock
Estima


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

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

From: "VIJAYA SARATHI N." <SARATHI@FPMLAN.iimahd.ernet.in>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: market integration
Date: Mon, 2 Nov 1998 23:15:45 GMT+0530
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:  IIM, Ahmedabad
X-mailer: Pegasus Mail v3.31 (via Mercury MTS (Bindery) v1.40)

Dear colleagues,

I'm trying to measure the extent of integration between the bond 
market and the stock market, through a VAR model.  I've yield curve 
data and the stock index. I do not have the price series for the bond 
market. I've only the yield series (constant maturity). Is it correct 
to use bond yields and the return on the stock index to infer market 
integration between the stock market and the bond market? Could you 
suggest to me, more appropriate functional form for the variables.

Your thoughts will help me a great deal. Thanx a lot in advance for 
your help.

Sarathi
Indian Institute of Management
Ahmedabad, India.

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

From: Hung-Jen Wang <hjwang@ieas.econ.sinica.edu.tw>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: a switch between OLS and 2SLS
Date: Tue, 03 Nov 1998 10:36:50 +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: QUALCOMM Windows Eudora Light Version 3.0.5 (32) (via Mercury MTS (Bindery) v1.40)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"

At 09:10 AM 11/2/98 -0500, you wrote:
>Why not just include the linreg command within the if-then-else statement?
>Or, if you have stuff to do in the meanwhile, just repeat the if, switching
>on method?
>
>Kit Baum
>

Because the example code I showed in the posting is a very simplified
version of the real one, and the real code already have many different
regression equations and also many levels of -if/else- statements used to
select different modeling.  Adding another level of -if/else- is somthing I
wish to avoid, if possible.  So, could someone help?

H.J. Wang




>--On Mon, Nov 2, 1998 19:42 +0800 "Hung-Jen Wang"
><hjwang@ieas.econ.sinica.edu.tw> wrote: 
>
>> Hello,
>> 
>>      I'd like to ask a hopefully simple programming problem.  In my
>> program I wish to let users have an option of choosing between OLS
>> estimation and 2SLS (IV).  Is there a simple way to implement such a
>> "switch"? Ideally, the program should look like the following:
>> 
>> ===========
>> 
>> <read in data and so on>
>> 
>>    ** Customization Section **
>> compute method = 1   ; * 1: OLS, 2: IV
>> .
>> .
>>    ** end of customization **
>> .
>> .
>> <skip main body>
>> .
>> .
>>    ** begin estimation **
>> 
>> if method ==2   ; * that is, you choose to do IV
>>   {
>>   compute [string] widget = 'inst'
>>   instruments z1 z2 z3 z4
>>   }
>> else
>>   {
>>   compute [string] widget = ' '
>>   }
>> end if
>> 
>> 
>> linreg(widget) y
>> # x1 x2 x3 x4
>> 
>> ======================
>> 
>> It doesn't work because of the string variable "widget".  Rates
>> returns something like this:
>> 
>>      linreg(widget)<<<<
>>      ## OP3. This Instruction Does Not Have An Option wid
>>      >>>>
>> 
>> Any suggestion toward solving the problem will be appreciated!
>> 
>> 
>> 
>> --------------------------------------------------------------------------
>> Hung-Jen Wang                     internet: hjwang@ieas.econ.sinica.edu.tw
>> Institute of Economics            Phone: 886-2-26533595 ext 428           
> 
>> Academia Sinica                   Fax:   886-2-27853946 
>> Taipei 115, Taiwan
>> --------------------------------------------------------------------------
>> 
>> 
>
>
>


--------------------------------------------------------------------------
Hung-Jen Wang                     internet: hjwang@ieas.econ.sinica.edu.tw
Institute of Economics            Phone: 886-2-26533595 ext 428             
Academia Sinica                   Fax:   886-2-27853946 
Taipei 115, Taiwan
--------------------------------------------------------------------------



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

From: "Maurice A. Harris" <maharr17@mailbox.syr.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Correction for Non-synchronous Trading in levels
Date: Tue, 03 Nov 1998 00:04:39 -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>
Organization: Finance Department, Syracuse University
X-Mailer: Mozilla 4.0 [en] (Win95; U) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit


I am interested in locating a RATS routine which implements the
Jokivuolle (JFQA, 1995)
procedure to purge a time series of non-synchronous trading effects.
Since I
ultimately need to have the "corrected series" in levels rather than
returns (as in Stoll and Whaley), the Jokivuolle approach seems to be
more
appropriate.

Any suggetions/comments or sample programs would be greatly appreciated.





---------- 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:_Kalman_filter?=
Date: Tue, 3 Nov 1998 11:30:53 +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) v1.40




I read it but it is not clear how i can implement my problem with Rats.
I am trying to estimate a system like :
X(t) = alpha + beta*F(t) + e(t)
F(t) = AF(t-1)+eta(t)
where V(e(t)) = Identity, v(eta(t))=Identity, F is of dimension 2 and
E{e(t)eta(t-k)}=0, for all k.
Jacques.



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

From: Jose Angelo Costa do Amor Divino <jacosta@guarany.cpd.unb.br>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Routine HISTORY
Date: Wed, 4 Nov 1998 16:27:00 -0300 (GRNLNDST)
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Mailer: Mercury MTS (Bindery) v1.40


Hello,

I'm working with structural chocks and I'm wearing the routine HISTORY of 
RATS. But, I need same methodological information about it and I'm not 
finding. So, I'd like to know if somebody could help me by indicating 
some study which describe it. The only reference that I have is Prof. Cris 
Sims' paper:
 "Comparison of interwar and postwar business cycles: monetarism 
reconsidered", American Economic Review, May 1980, pp. 250-257. 
However, in this paper, He only wear Variance Decomposition and Impulse 
Response Functions. He don't wear the HISTORY.

Since now, I thanks.
Sincerely,
				J. Angelo.





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

From: Ashley Lyman <alyman@uidaho.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: conditioning diagnostics
Date: Wed, 4 Nov 1998 12:38:08 -0800 (PST)
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Mailer: Mercury MTS (Bindery) v1.40

I am interested in any rats procedures that may have been developed to
yield conditioning diagnostics focussing on collinearity and weak data in
regression.  Specifically the reference is a 1991 book (Wiley) by David A.
Belsley in which he discusses and develops various diagnostic statistics
for assessing collinearity for linear models and for models with
logarithms. Also, are there any procedures available that would perform
singular-value decomposition of a matrix? I am trying to find out if I
have to do the programming in Rats myself or whether someone else has
already done it.

Let me thank you in advance for any assistance you can provide.
	
regards,

Ashley Lyman
Department of Economics
Univ. of Idaho
Moscow, Id. 83844

208-885-7145  office
alyman@uidaho.edu 




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

From: Michael Hanson <mhanson@econ.lsa.umich.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: OVERLAY usage (or, NT <-> Unix bug?)
Date: Thu, 5 Nov 1998 01:10:24 -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: MULTIPART/MIXED; BOUNDARY="-559023410-851401618-910245532=:17161"
X-Mailer: Mercury MTS (Bindery) v1.40

  This message is in MIME format.  The first part should be readable text,
  while the remaining parts are likely unreadable without MIME-aware tools.
  Send mail to mime@docserver.cac.washington.edu for more info.

---559023410-851401618-910245532=:17161
Content-Type: TEXT/PLAIN; CHARSET=US-ASCII
Content-ID: <Pine.SOL.3.95.981105010357.17405B@smith.econ.lsa.umich.edu>

	I am getting a very odd error from using the OVERLAY command in a
procedure file, and I'd appreciate it if anyone can point out my error. 
(It is entirely possible that my syntax is incorrect.)  However, here's
the kicker:  this procedure ran fine on a Windows NT machine, but returns
an error with the Unix version!!  (The Unix version is 4.30;  I believe
the PC version was 4.3 as well, but I no longer have access to that
machine -- nor any other PC with RATS.  I reloaded the files from an
archive of the NT machine, so I don't believe the code (or the program
file which calls it) has been altered between the two experiments.)

	I have attached the complete procedure to this message, but the
block of code where the error arises is as follows (starting at line 46): 

        BOXJENK(AR=slag,MA=1,NOPRINT) sdiff startl+slag endl
        
        DIM phi(slag)
        OVERLAY %BETA(1) WITH phi(slag)
        COMP phi = phi*(-1.0)

        FILTER series startl+slag endl ystar
        # 1 TO slag
        # phi

The following error message occurs when SOURCING the procedure (i.e. 
*not* upon execution but merely loading the procedure into memory),
immediately after the line with the OVERLAY command: 

## SX22. Expected Type EQUATION, Got REAL Instead

Given the discussion of the OVERLAY command, I have *no idea* why RATS
"expects type EQUATION"!  Shouldn't it "expect" two arrays?  (I told you I
was uncomfortable with the syntax...) 


	What the attached procedure is *supposed* to do is estimate the
Leybourne-McCabe (1994 JBES) unit root test.  Unfortunately, since it
won't run I can't debug it.  (Yes, I know the lag length should be chosen
by AIC rather than SIC -- that's one of the things I was planning to fix.)

	I would appreciate it if anyone could provide one or more of the
following: (a) an explanation of what I'm doing wrong (code-wise) and how
to fix it, (b) suggestions for an alternative (better?) way to do the same
thing, (c) confirmation that the procedure works on non-Unix platforms,
and/or (d) an already written and debugged procedure that performs this
test.  Any enterprising soul who wishes to take this code and rewrite it
should feel free to do so;  I would just ask for a copy be sent to me and
some credit in the procedure source code (if appropriate).  Any other
constructive comments on the procedure are welcome.  Thanks! 

					-- Mike

P.S. Use this procedure at your own risk!  No implied or expressed 
warrantee, yadda, yadda, yadda....

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

---559023410-851401618-910245532=:17161
Content-Type: TEXT/PLAIN; CHARSET=US-ASCII; NAME="lmurs.src"
Content-Transfer-Encoding: BASE64
Content-ID: <Pine.SOL.3.95.981105005852.17161C@smith.econ.lsa.umich.edu>
Content-Description: LMURS.SRC
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---559023410-851401618-910245532=:17161--

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

From: "Estima" <estima@estima.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: OVERLAY usage (or, NT <-> Unix bug?)
Date: Thu, 5 Nov 1998 10:36:17 -0600
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
X-mailer: Pegasus Mail for Win32 (v2.54) (via Mercury MTS (Bindery) v1.40)

>
> <snip>
> The following error message occurs when SOURCING the procedure (i.e. 
> *not* upon execution but merely loading the procedure into memory),
> immediately after the line with the OVERLAY command: 
> 
> ## SX22. Expected Type EQUATION, Got REAL Instead
> 
 

Just to let everyone know, this problem with the OVERLAY is a 
UNIX-specific bug. If anyone else runs into it, let us know and we'll 
get you a fix.

Sincerely,
Tom Maycock
Estima


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

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

From: Jose Angelo Costa do Amor Divino <jacosta@guarany.cpd.unb.br>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Routine HISTORY
Date: Thu, 5 Nov 1998 12:50:16 -0300 (GRNLNDST)
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Mailer: Mercury MTS (Bindery) v1.40


I'm working with structural chocks and I'm wearing the routine HISTORY of 
RATS. But, I need same methodological information about it and I'm not 
finding. So, I'd like to know if somebody could help me by indicating 
some study which describe it. The only reference that I have is Prof. Cris 
Sims' paper:
 "Comparison of interwar and postwar business cycles: monetarism 
reconsidered", American Economic Review, May 1980, pp. 250-257. 
However, in this paper, He only wear Variance Decomposition and Impulse 
Response Functions. He don't wear the HISTORY.

Since now, I thanks.
Sincerely,
				J. Angelo.

Department of Economics
University of Brasilia - Brazil

jacosta@guarany.cpd.unb.br







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

From: mark.astley@bankofengland.co.uk (Mark Astley)
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Routine HISTORY
Date: Fri, 06 Nov 1998 09:46:53 +0000
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
X-Mailer: Mozilla 4.05 [en] (WinNT; I) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Try:

Burbridge and Harrison (1985) "A Historical Decomposition of the Great
Depression to Determine the Role of Money", Journal of Monetary Economics, 16,
pp 45-54.

MSA


Jose Angelo Costa do Amor Divino wrote:

> I'm working with structural chocks and I'm wearing the routine HISTORY of
> RATS. But, I need same methodological information about it and I'm not
> finding. So, I'd like to know if somebody could help me by indicating
> some study which describe it. The only reference that I have is Prof. Cris
> Sims' paper:
>  "Comparison of interwar and postwar business cycles: monetarism
> reconsidered", American Economic Review, May 1980, pp. 250-257.
> However, in this paper, He only wear Variance Decomposition and Impulse
> Response Functions. He don't wear the HISTORY.
>
> Since now, I thanks.
> Sincerely,
>                                 J. Angelo.
>
> Department of Economics
> University of Brasilia - Brazil
>
> jacosta@guarany.cpd.unb.br




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

From: Price SG <S.G.Price@city.ac.uk>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: 
Date: Fri, 6 Nov 1998 19:08:47 +0000 (BST)
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Mailer: Mercury MTS (Bindery) v1.40

Dear RATS users -

I am trying to loop over forecasts with different sample periods.

Running the loop explicitly so you can see what's going on,  I get 

**************************************
 compute ss=1976:4
 smpl ss ss+20
 forecast(model=input)   14

[which works fine]

 compute ss=1977:4
 smpl ss ss+20
 forecast(model=input)   14
## FO1. Equations 1 and 1 Both Have Dependent Variable LLY

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

RATS seems to think I have two equations in the model with the same dv.

Can anyone help?






*******************************************************************
 Prof Simon Price         email:      s.g.price@city.ac.uk 
 Department of Economics
 City University          Tel (work): +44 (0)171 477 8000 ext 4503
 Northampton Square       Fax (work): +44 (0)171 477 8580
 London EC1V 0HB UK       
 URL http://www.city.ac.uk/~sm344
*******************************************************************





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

From: Ashley Lyman <alyman@uidaho.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Programming in Rats
Date: Fri, 6 Nov 1998 16:21:01 -0800 (PST)
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Mailer: Mercury MTS (Bindery) v1.40

	I would like to iterate over a set of data series (my example
below has just two series-X and Y--assume 120 obs).  I specifically want
to multiply each element of series by a number (computed by taking the
product of all elements of the series, raising the product to the power
1/120 and dividing the result into exp(1))

-----------------------------------
allocate 120
open data c:\mydat.rat
data(format=rats)

* X and Y are two data series with 120 obs. each

dofor [series] I = X Y
clear gg
set [series] gg =I
compute product=1.0
do J=1,120
   compute product=product*gg(J)
end do
set [series] I =log(gg*exp(1)/(product**(1/120)))
end do I
--------------------


unfortunately the above procedure results in X and Y ending up
with different but constant values across observations.

What am I doing wrong?  Also, is there a source for more explanation on
programing in Rats?


R. Ashley Lyman
Professor of Economics
Department of Economics
Univ. of Idaho
Moscow, Idaho 83844
208-885-7145



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

From: KHL2288@UTARLG.UTA.EDU
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Likelihood Ratio Test in Bivariate GARCH
Date: Sun, 08 Nov 1998 12:14:57 -0600 (CST)
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-version: 1.0
Content-type: TEXT/PLAIN; CHARSET=US-ASCII
X-Mailer: Mercury MTS (Bindery) v1.40

Dear RATS users,
I'd like to know how to conduct likelihood ratio test in Bivariate GARCH. I am using the sample program written by Rob Trevor. For example, if I have the following:
NONLIN B11 B12 B21 B22
FRML RESID1 = Y1 - B11 - B12*X1
FRML RESID2 = Y2 - B21 - B22*X2

And, I'd like to conduct LR test for the null of B22=0.
Thank you so much for your help.

Best regards,

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

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

From: Luis Fernando =?iso-8859-1?Q?Mej=EDa?= Alzate <lu-mejia@uniandes.edu.co>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: ARIMA models and seasonal adjustment
Date: Mon, 09 Nov 1998 00:16:36 -0500
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Organization: Universidad de los Andes
X-Mailer: Mozilla 4.5b2 [en] (Win95; I) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable

I need help on how to estimate these models using the BOXJENK
instruction (or any other similar):
ARIMA(1,1,1) x ARIMA_4(1,0,0) where the ARIMA_4(1,0,0) is the stationary
model to estimate. I also need to estimate a ARIMA(1,1,1) x
ARIMA_6(1,1,0) x ARIMA_12(1,1,0) model. Is there any way I can do this
using standard RATS procedure?

Thanks in advance for your help.

Luis Fernando Mej=EDa Alzate
Facultad de Econom=EDa
Universidad de los Andes
Santaf=E9 de Bogot=E1, D. C. - Colombia



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

From: "Frieder Knüpling" <knueplin@ruf.uni-freiburg.de>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: SCATTER
Date: Mon, 09 Nov 1998 18:54:15 +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: Mozilla 4.06 [en] (Win98; I) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Dear RATS users,

I wonder wether there is any way to produce SCATTER-plots that use
different styles for different x-y-pairs (symbols/lines).

Yours, Frieder

--
Frieder Knuepling
Albert-Ludwigs-Universitaet Freiburg
Institut fuer Allgemeine Wirtschaftsforschung
Abteilung Statistik und Oekonometrie
Belfortstr. 24
D-79098 Freiburg
Tel +49 761 / 203 - 2341
Fax +49 761 / 203 - 2340



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

From: wobauer@studi.unizh.ch
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Seasons
Date: Mon, 9 Nov 1998 22:18:06 +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
Content-transfer-encoding: 7BIT
X-mailer: Pegasus Mail for Win32 (v3.01b) (via Mercury MTS (Bindery) v1.40)

Dear RATS users,

I would like to do seasonal analysis (mainly calculation of seasonal 
indices) of daily data. The seasons should be of length 1 week, 2 
weeks, 1 month, 1 quarter of a year and so on. Has anybody of you 
experience with determining the respective period of an 
obeservation? E.g. which week 1998:08.25 was?

Further: what is the most suitable representation of the data? A 
rect[serires] with one row equal to one year, every entry being a 
period?

I am grateful for any hints, as I am just starting thje project.

Greetings

Wolfgang Bauer

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

From: Klaus Fischer <Klaus.Fischer@fas.ulaval.ca>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Confidence bands in forecasts with simultaneous equations
Date: Tue, 10 Nov 1998 10:47:16 -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>
Organization: CREFA-Université Laval
X-Mailer: Mozilla 4.01 [en] (Win95; I) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="------------69A2DF769DFA44AB6CB480D5"

This is a multi-part message in MIME format.
--------------69A2DF769DFA44AB6CB480D5
Content-Type: text/plain; charset=iso-8859-1
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit

I'm a PhD student and I have a problem with RATS.

My problem is to obtain the confidence bands (interval forecasts) of
punctual forecasts when we have simultaneous equations. I've got the
punctual forecasts (for y1 and y2) but I couldn't find the instructions
about the confidence bands of this punctual forecasts. In others words,
I want the confidence bands associate to each punctual forecasts of
endogenous variables of the system. (I've found it for the VAR context
but not for the simultaneous equations).

Next, I present the RATS program that I used.

**********************************************************
allocate 250
open data c:\DONNEES
data(format=wks,org=obs) 1 250 x1 x2 x3 y1 y2
*
* x1 isn't included into the simultaneous equations
* x2 and x3 are the exogenous variables
* y1 and y2 are the endogenous variables
*
INSTRUMENTS constant y1{0 5} y2{0 to 2} x2{1 2} x3{1 2}
ar1(inst,frml=y1eq) y1
# constant y1{5} y2{1 to 2} x2{1} x3{2}
ar1(inst,frml=y2eq) y2
# constant y2{2} x2{2} x3{1}
frml(identity) x2id x2 = x2
frml(identity) x3id x3 = x3
GROUP reduit y1eq>>F_y1  y2eq>>F_y2 x2id x3id
FORECAST(model=reduit, print) 2 25
**********************************************************

If anyone can give me a hint about how to do it, I will appreciate to
receive your comments.

Sincerly,

Ginette Bégin
Laval University
Québec (Canada)
e-mail : aaf155@agora.ulaval.ca


--------------69A2DF769DFA44AB6CB480D5
Content-Type: text/x-vcard; charset=us-ascii; name="vcard.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Klaus Fischer
Content-Disposition: attachment; filename="vcard.vcf"

begin:          vcard
fn:             Klaus Fischer
n:              Fischer;Klaus
org:            CREFA - Université Laval
adr:            ;;;Quebec (PQ);;G1K 7P4;Canada
email;internet: Klaus.Fischer@fas.ulaval.ca
tel;work:       418-656-2131  X3679
tel;fax:        418-656-7746
x-mozilla-cpt:  ;0
x-mozilla-html: FALSE
end:            vcard


--------------69A2DF769DFA44AB6CB480D5--


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

From: Frederick Bourgoin <bourgoif@ubkam.com> (by way of Rob Trevor)
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Programming in Rats
Date: Wed, 11 Nov 1998 09:06:00 +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
Content-Type: text/plain; charset="us-ascii"
X-Mailer: Mercury MTS (Bindery) v1.40


allocate 120
open data c:\mydat.rat
data(format=rats)


* X and Y are two data series with 120 obs. each


dofor [series] I = X Y
clear gg
set [series] gg =I
compute product=1.0
do J=1,120
   compute product=product*gg(J)
end do
set [series] I =log(gg*exp(1)/(product**(1/120)))
end do I


You Should write probably :

set [series] I =log(gg{0}*exp(1)/(product**(1/120)))

Frederick Bourgoin

---------------------------------------------------------------
The views expressed are of the individual, and do not
necessarily reflect the views of The United Bank of Kuwait PLC.
---------------------------------------------------------------

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

From: "Estima" <estima@estima.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Programming in Rats
Date: Tue, 10 Nov 1998 17:04:52 -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) v1.40)

> 
> dofor [series] I = X Y
> clear gg
> set [series] gg =I
> compute product=1.0
> do J=1,120
>    compute product=product*gg(J)
> end do
> set [series] I =log(gg*exp(1)/(product**(1/120)))
> end do I
> 
> 
> You Should write probably :
> 
> set [series] I =log(gg{0}*exp(1)/(product**(1/120)))
> 

I already answered this privately, but the group may benefit from 
more info, as the suggestion above won't help and because there are 
actually 2 problems here, as described below. 

The problem is that if you do:


dofor [series] I = X Y


I will be an integer, not a series (the [series] tag has no effect 
here because I is a reserved variable, of type integer, which is used 
in the EWISE instruction). As a result, I will contain the integer 
series number for the current series in the list.

That's fine, and in fact we recommend that you allow the DOFOR index 
to be an integer (which is what happens by default if you haven't 
declared the index variable ahead of time), becuase declaring it as a 
series can lead to other problems.

However, you have to be careful when you refer to "I" in a context 
where RATS isn't necessarily expecting a series, such as:

 set [series] gg =I

(note that the [series] tag here is completely superfluous--the first 
parameter in a SET instruction must always be a series, so the tag 
has no effect).

The problem is that there's no reason why the "I" on the right 
hand side of the = can't be interpreted as an integer. So, you need 
to tell RATS to treat it as a series number, by doing:

set gg = I{0}

or

set gg = ([series] I )

The other problem is with the line:

set [series] I =log(gg*exp(1)/(product**(1/120)))

Note that here again, the [series] tag is superfluous. The first 
parameter in a SET must be a series, so RATS automatically treats I 
as being a series number. The problem is actually with the 
expression:

(1/120)

In RATS, divding two integers gives you an integer result (any 
remainder is truncted). So, this expression returns zero. To get the 
fractional value, make either or both parts real. For example, 
replace:

set I =log(gg*exp(1)/(product**(1/120)))

with:

set I =log(gg*exp(1)/(product**(1.0/120)))

See page 1-31 in the manual for details.

Sincerely,
Tom Maycock
Estima

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

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

From: Arturo Galindo <agalindo@students.uiuc.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Markov-Switching Models.
Date: Thu, 12 Nov 1998 15:29:39 -0600
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
X-Mailer: Mozilla 4.01 [en] (Win95; I) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

> As there's been a lot of interest in these models lately, we've
> decided to try and put together some Markov switching model examples.
> I hope to be able to post something in the next day or two that may
> be of help.
>
> Also, we're planning on putting out an issue of our newsletter in the
> next few weeks, which will include further information on Markov
> models. As part of our work on Version 5, we're also looking into
> making some improvements that will make handling these models easier
> and extend the range of models that can be handled.
>
> Sincerely,
> Tom Maycock
> Estima
>
> ------------------------------------------------------------
> |   Estima                    |  Sales:   (800) 822-8038   |
> |   P.O. Box 1818             |  Support: (847) 864-1910   |
> |   Evanston, IL 60204-1818   |  Fax:     (847) 864-6221   |
> |   USA                       |  estima@estima.com         |
> |                             |  http://www.estima.com     |
> ------------------------------------------------------------

Dear Tom.

I just wanted to know if there is an update about the newsletter related
to Markov-Switching models.

In the mean time, does anyone have a procedure  to estimate these? I
would truly appreciate if someone helps me to get hold of it.

Thanks,

Arturo Galindo
Department of Economics
University of Illinois at Urbana-Champaign


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

From: Simon.Van-Norden@hec.ca
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Markov-Switching Models.
Date: Thu, 12 Nov 1998 16:58:16 -0500
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Organization: Ecole des Hautes Etudes Commerciales
X-Mailer: Mozilla 4.5 [en] (Win95; U) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit

I've got lots of code on my home page, but its all in GAUSS.
(Ditto for James Hamilton.)

Arturo Galindo wrote:
> 
> > As there's been a lot of interest in these models lately, we've
> > decided to try and put together some Markov switching model examples.
> > I hope to be able to post something in the next day or two that may
> > be of help.
> >
> > Also, we're planning on putting out an issue of our newsletter in the
> > next few weeks, which will include further information on Markov
> > models. As part of our work on Version 5, we're also looking into
> > making some improvements that will make handling these models easier
> > and extend the range of models that can be handled.
> >
> > Sincerely,
> > Tom Maycock
> > Estima
> >
> > ------------------------------------------------------------
> > |   Estima                    |  Sales:   (800) 822-8038   |
> > |   P.O. Box 1818             |  Support: (847) 864-1910   |
> > |   Evanston, IL 60204-1818   |  Fax:     (847) 864-6221   |
> > |   USA                       |  estima@estima.com         |
> > |                             |  http://www.estima.com     |
> > ------------------------------------------------------------
> 
> Dear Tom.
> 
> I just wanted to know if there is an update about the newsletter related
> to Markov-Switching models.
> 
> In the mean time, does anyone have a procedure  to estimate these? I
> would truly appreciate if someone helps me to get hold of it.
> 
> Thanks,
> 
> Arturo Galindo
> Department of Economics
> University of Illinois at Urbana-Champaign

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

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

From: James Peery Cover <jcover@bama.ua.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Federal Funds Rate
Date: Thu, 12 Nov 1998 16:33:53 -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>
Organization: University of Alabama
X-Mailer: Mozilla 4.05 [en] (Win95; I) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Does anyone on this list know of a source where I can get time-series
data on the targeted value of the Federal Funds rate?  (I know I can get
the value of the rate, but I am looking for the target value.)
-- 
James Peery Cover
Department of Economics and Finance
Culverhouse College of Commerce
University of Alabama
P. O. Box 870224
Tuscaloosa, AL 35487-0224

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

From: "Grahl, Paulo" <paulo.grahl@csfb.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: RE: Federal Funds Rate
Date: Thu, 12 Nov 1998 18:07:42 -0500
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2232.9) (via Mercury MTS (Bindery) v1.40)
Content-Type: text/plain

James,

Do you have access to Bloomberg services ?
If so, just type FDTR Index <GO> and you'll have access to daily date since
1985.


Regards,


--------------------------------------------
PAULO GRAHL, CFA
Phone: (55 11) 821 6346
Fax:     (55 11) 821 6908
e-mail: paulo.grahl@csfb.com 
Emerging Markets Research
Credit Suisse First Boston Garantia
--------------------------------------------

> -----Original Message-----
> From:	James Peery Cover [SMTP:jcover@bama.ua.edu]
> Sent:	Thursday, November 12, 1998 7:34 PM
> To:	RATS Discussion List
> Subject:	Re: Federal Funds Rate
> 
> Does anyone on this list know of a source where I can get time-series
> data on the targeted value of the Federal Funds rate?  (I know I can get
> the value of the rate, but I am looking for the target value.)
> -- 
> James Peery Cover
> Department of Economics and Finance
> Culverhouse College of Commerce
> University of Alabama
> P. O. Box 870224
> Tuscaloosa, AL 35487-0224

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

From: "Christopher F Baum" <baum@bc.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Federal Funds Rate
Date: Thu, 12 Nov 1998 18:31:35 -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) [1.3.4, s/n P020-300786-009] (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Even without Bloomberg, it's in Glenn Rudebusch's article, Journal of
Monetary Econ? Not sure of exact cite, but not hard to find in EconLit.

Kit Baum

--On Thu, Nov 12, 1998 16:33 -0600 "James Peery Cover" <jcover@bama.ua.edu>
wrote: 

> Does anyone on this list know of a source where I can get time-series
> data on the targeted value of the Federal Funds rate?  (I know I can get
> the value of the rate, but I am looking for the target value.)
> -- 
> James Peery Cover
> Department of Economics and Finance
> Culverhouse College of Commerce
> University of Alabama
> P. O. Box 870224
> Tuscaloosa, AL 35487-0224



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

From: phsu@mail.stit.edu.tw
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Kolmogorov-Smirnov D statistic
Date: Fri, 13 Nov 1998 08:51:01 +0000
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
X-mailer: Pegasus Mail for Windows (v2.41) (via Mercury MTS (Bindery) v1.40)

Dear RATS users,
  Could anyone tell me where I can find the procedure of 
Kolmogorov-Smirnov D statistic? Thanks a lot.
Philip Hsu

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

From: James Peery Cover <jcover@bama.ua.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Federal Funds Rate
Date: Fri, 13 Nov 1998 09:55:32 -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>
Organization: University of Alabama
X-Mailer: Mozilla 4.05 [en] (Win95; I) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Dear Paulo,
Thank you very much.  I think our library has access to Bloomberg.

Grahl, Paulo wrote:
> 
> James,
> 
> Do you have access to Bloomberg services ?
> If so, just type FDTR Index <GO> and you'll have access to daily date since
> 1985.
> 
> Regards,
> 
> --------------------------------------------
> PAULO GRAHL, CFA
> Phone: (55 11) 821 6346
> Fax:     (55 11) 821 6908
> e-mail: paulo.grahl@csfb.com
> Emerging Markets Research
> Credit Suisse First Boston Garantia
> --------------------------------------------
> 
> > -----Original Message-----
> > From: James Peery Cover [SMTP:jcover@bama.ua.edu]
> > Sent: Thursday, November 12, 1998 7:34 PM
> > To:   RATS Discussion List
> > Subject:      Re: Federal Funds Rate
> >
> > Does anyone on this list know of a source where I can get time-series
> > data on the targeted value of the Federal Funds rate?  (I know I can get
> > the value of the rate, but I am looking for the target value.)
> > --
> > James Peery Cover
> > Department of Economics and Finance
> > Culverhouse College of Commerce
> > University of Alabama
> > P. O. Box 870224
> > Tuscaloosa, AL 35487-0224

-- 
James Peery Cover
Department of Economics and Finance
Culverhouse College of Commerce
University of Alabama
P. O. Box 870224
Tuscaloosa, AL 35487-0224

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

From: Mark Tomsett <M.P.Tomsett@durham.ac.uk>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: State-Space Modelling.
Date: Tue, 17 Nov 1998 15:09:28 +0000 (GMT)
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Mailer: Mercury MTS (Bindery) v1.40

Dear All,

Does anyone know of any sources of code to run a local linear trend state
space model?

Any help would be much appreciated.

Yours,

Mark.

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

Mark Tomsett
Department of Economics
University of Durham
23-26 Old Elvet
DURHAM CITY
County Durham

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



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

From: Ying-Man Chung <y_mchung@yahoo.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: RE:code for explosive root test
Date: Tue, 17 Nov 1998 07:40:40 -0800 (PST)
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mercury MTS (Bindery) v1.40

Dear RATS users,

I wish to request for codes for unit root testing against explosive
alternatives. I know that ursb.src can implement Bhargava's R1 and R2
tests. However, I want to implement his N1 and N2 statistics. May I
ask any codes to implement the above statistics, please?? Thank you
very much.



REgards,

Sarah
HKSYC 
 




_________________________________________________________
DO YOU YAHOO!?
Get your free @yahoo.com address at http://mail.yahoo.com


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

From: Simon.Van-Norden@hec.ca
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: State-Space Modelling.
Date: Tue, 17 Nov 1998 12:21:24 -0500
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Organization: Ecole des Hautes Etudes Commerciales
X-Mailer: Mozilla 4.5 [en] (Win95; U) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit

I looked for some over the summer without success and had to switch to GAUSS.

If you find some for this and other related models, could you please post it to
the list?

(I get the impression that the KALMAN command and its documentation are
sufficiently cryptic that they have really discouraged people from developing
such procedures.)

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

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

From: Simon.Van-Norden@hec.ca
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Bhargava
Date: Tue, 17 Nov 1998 15:03:19 -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>
Organization: Ecole des Hautes Etudes Commerciales
X-Mailer: Mozilla 4.5 [en] (Win95; U) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: multipart/mixed;

This is a multi-part message in MIME format.
--------------1FEF7A8C2130C4C1D3B8C81C
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit

I've got some *old* code that I'll post which does the R1, R2, N1 and N2
Bhargava statistics.  (It's been so long since I've used it that I no longer
recall what the differences are.)
-- 
Simon van Norden			Professeur invité
E-mail:  simon.van-norden@hec.ca   or    svn@alum.mit.edu 
Home Page:       http://www.hec.ca/pages/simon.van-norden
(514)340-6781 or fax (514)340-5632 or  cell (514)993-0466
Service de l'enseignement de la finance, École des H.E.C.
3000 Cote-Sainte-Catherine, Montreal QC, CANADA   H3T 2A7
--------------1FEF7A8C2130C4C1D3B8C81C
Content-Type: application/x-unknown-content-type-src_auto_file;
 name="bhargava.src"
Content-Transfer-Encoding: base64
Content-Disposition: inline;
 filename="bhargava.src"
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=
--------------1FEF7A8C2130C4C1D3B8C81C--


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

From: "Estima" <estima@estima.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: State-Space Modelling.
Date: Tue, 17 Nov 1998 14:45:47 -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
Content-transfer-encoding: 7BIT
X-mailer: Pegasus Mail for Win32 (v2.54) (via Mercury MTS (Bindery) v1.40)

> Does anyone know of any sources of code to run a local linear trend state
> space model?
> 

Mark:

Don't know for sure--but it sounds like this would something you
could do with a general Kalman filtering routine, or some variation
thereof. See, for example, the KFILTER.SRC procedure available on
our web site. If KFILTER doesn't do the trick exactly as written, I
suspect it could be fairly easily modified to do the job.

BTW, note that the KALMAN instruction (and the related instructions)
are quite powerful, but they were designed to handle only a fairly
specific subset of models (basically where the state variables are
regression coefficients). More general state space modeling tasks 
require a different approach--the KFILTER.SRC proc, for example.

Sincerely,
Tom Maycock
Estima


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

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

From: "Alex Hinder" <ahinder@swissonline.ch>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Markowitz Mean/Variance Portfolio Optimization Method
Date: Wed, 18 Nov 1998 10:19:01 +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: multipart/alternative;
X-Mailer: Microsoft Outlook Express 4.71.1712.3 (via Mercury MTS (Bindery) v1.40)

This is a multi-part message in MIME format.

------=_NextPart_000_0025_01BE12DC.DCB977C0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

I am looking for a RATS application of the Markowitz portfolio =
optimization method to the optimal allocation of stocks and securities =
in a portfolio. Or has somebody already experimented with even more =
sophisticated applications (f.ex. allowing for transaction costs)?


Alex Hinder
ahinder@swissonline.ch

------=_NextPart_000_0025_01BE12DC.DCB977C0
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

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

<META content=3Dtext/html;charset=3Diso-8859-1 =
http-equiv=3DContent-Type>
<META content=3D'"MSHTML 4.71.1712.3"' name=3DGENERATOR>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT color=3D#000000 size=3D2>I am looking for a RATS application =
of the=20
Markowitz portfolio optimization method to the optimal allocation of =
stocks and=20
securities in a portfolio. Or has somebody already experimented with =
even more=20
sophisticated applications (f.ex. allowing for transaction =
costs)?</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT color=3D#000000 size=3D2>Alex Hinder</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2><A=20
href=3D"mailto:ahinder@swissonline.ch">ahinder@swissonline.ch</A></FONT><=
/DIV></BODY></HTML>

------=_NextPart_000_0025_01BE12DC.DCB977C0--


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

From: "Alex Hinder" <ahinder@swissonline.ch>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Markowitz Mean/Variance Portfolio Optimization Method
Date: Thu, 19 Nov 1998 08:26:02 +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: multipart/alternative;
X-Mailer: Microsoft Outlook Express 4.71.1712.3 (via Mercury MTS (Bindery) v1.40)

This is a multi-part message in MIME format.

------=_NextPart_000_000D_01BE1396.3E66CFE0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

I am looking for a RATS application of the Markowitz portfolio =
optimization method to the optimal allocation of stocks and securities =
in a portfolio. Or has somebody already experimented with even more =
sophisticated applications (f.ex. allowing for transaction costs)?
=20
=20
Alex Hinder
ahinder@swissonline.ch

------=_NextPart_000_000D_01BE1396.3E66CFE0
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

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

<META content=3Dtext/html;charset=3Diso-8859-1 =
http-equiv=3DContent-Type>
<META content=3D'"MSHTML 4.71.1712.3"' name=3DGENERATOR>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV>
<DIV><FONT color=3D#000000 size=3D2>I am looking for a RATS application =
of the=20
Markowitz portfolio optimization method to the optimal allocation of =
stocks and=20
securities in a portfolio. Or has somebody already experimented with =
even more=20
sophisticated applications (f.ex. allowing for transaction =
costs)?</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT color=3D#000000 size=3D2>Alex Hinder</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2><A=20
href=3D"mailto:ahinder@swissonline.ch">ahinder@swissonline.ch</A></FONT><=
/DIV></DIV></BODY></HTML>

------=_NextPart_000_000D_01BE1396.3E66CFE0--


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

From: Winnie Mabaleka <WMabaleka@bank-banque-canada.ca>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Unit root test for panel data
Date: Fri, 20 Nov 1998 09:50:35 -0500
Errors-to: <jandrews@efs1.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 4.05 [en] (X11; U; SunOS 5.5.1 sun4u) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Hi Subscribers,

I am a research student still familiarizing myself with RATS and this
question may sound pretty mundane to most of you. I want to test a set
of panel data for unit root and it doesnt seem obvious from the coded
source programs that I have applied on a single time series. No one
around is familiar with the procedure for testing unit root on panel
data.
I would like to follow a procedure that was developed by Levin and Lin,
1993 (Discussion Paper # 93-56 Dept of Economics, University of
Carlifornia, San Diego).  Does anyone have a coded RATS program to do
this test? I would appreciate your help.

Thanks in advance for all your help.

Winnie Mabaleka
Organization: Bank of Canada / Banque du Canada
 maba@bank-banque-canada.ca


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

From: "Christopher F Baum" <baum@bc.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Unit root test for panel data
Date: Fri, 20 Nov 1998 10:01:55 -0500
Errors-to: <jandrews@efs1.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) [1.3.4, s/n P020-300786-009] (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Peter Pedroni of Indiana U has developed a number of panel unit root tests,
and he will make his RATS code available. I suggest you contact him
(ppedroni@indiana.edu). 

Kit Baum
Boston College

--On Fri, Nov 20, 1998 9:50 -0500 "Winnie Mabaleka"
<WMabaleka@bank-banque-canada.ca> wrote: 

> Hi Subscribers,
> 
> I am a research student still familiarizing myself with RATS and this
> question may sound pretty mundane to most of you. I want to test a set
> of panel data for unit root and it doesnt seem obvious from the coded
> source programs that I have applied on a single time series. No one
> around is familiar with the procedure for testing unit root on panel
> data.
> I would like to follow a procedure that was developed by Levin and Lin,
> 1993 (Discussion Paper # 93-56 Dept of Economics, University of
> Carlifornia, San Diego).  Does anyone have a coded RATS program to do
> this test? I would appreciate your help.
> 
> Thanks in advance for all your help.
> 
> Winnie Mabaleka
> Organization: Bank of Canada / Banque du Canada
>  maba@bank-banque-canada.ca
> 



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

From: Ying-Man Chung <y_mchung@yahoo.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Bhargava
Date: Sun, 22 Nov 1998 09:23:25 -0800 (PST)
Errors-to: <jandrews@efs1.efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mercury MTS (Bindery) v1.40






---Simon.Van-Norden@hec.ca wrote:

Dear Dr.Norden and other RATS users,

I've asked for RATs code for Bhargava's N1 and N2 test. and I received
Dr.Norden's reply but nothing's shown. Might I ask what did it mean
from Dr.Norden's letter? Any codes you want to give me? Thank you very
much!


Regards,

Sarah
_________________________________________________________
DO YOU YAHOO!?
Get your free @yahoo.com address at http://mail.yahoo.com


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

From: "Wolfgang Bauer" <wobauer@access.unizh.ch>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Accessing computer's time counter
Date: Tue, 24 Nov 1998 22:01:49 +0100
Errors-to: <jandrews@efs1.efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
X-mailer: Pegasus Mail for Win32 (v3.01b) (via Mercury MTS (Bindery) v1.40)

Dear Rats users,

I would like to fasten up some of my RATS procedures. Is there a 
way to get the value of an internal time counter to determine how 
much time it takes RATS to work through some line of code?

Thanks for hints, kind regards

Wolfgang Bauer

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

From: Fawzy Omran <M.F.Omran@ma.hw.ac.uk>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Specification tests
Date: Wed, 25 Nov 1998 14:29:34 GMT
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
X-Mailer: Mercury MTS (Bindery) v1.40


(1) I am looking for a Rats subroutine which can test for misspecification using the information matrix test proposed by White (1982) or Lancaster (1984).

Lancaster, Tony (1984), "the covariance matrix of the information matrix test," Econometrica, 52(4), 1051-1053.

White, Halbert (1982), "maximum likelihood estimation of misspecified models," Econometrica, 50(1), 1-25.


(2) subroutines to evaluate the goodness of fit of the t distribution and the generalised error distribution discussed in Nelson (1991).

Nelson, D B (1991), "conditional heteroscedasticity in asset returns: a new approach," Econometrica, 59(2), 347-70.


Many thanks


M F Omran
Dept. of Actuarial Maths/Stats
Heriot Watt University
Edinburgh EH14 4AS
Fax: 0131 451 3249
email: m.f.omran@ma.hw.ac.uk

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

From: hideakii <hideakii@yahoo.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Exponentiation
Date: Wed, 25 Nov 1998 12:25:04 -0800 (PST)
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mercury MTS (Bindery) v1.40

Dear RATS Users

I got differencial results using 2 expression as follows:

1. x**(1.5)
2. x**(3/2)

and latter results was not one which I expected.
Was it incorrect expression??
I have checked Manual(1-29,30), but I could not find explanations
related to this.

Thanks in advance

ISHIGAMI Hideaki
Nippon Sports Science Univ.
_________________________________________________________
DO YOU YAHOO!?
Get your free @yahoo.com address at http://mail.yahoo.com


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

From: "Wai Lee" <lee_wai@jpmorgan.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Exponentiation
Date: Wed, 25 Nov 1998 15:29:02 -0500
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Mime-Version: 1.0
Content-type: text/plain; charset=us-ascii
X-Mailer: Mercury MTS (Bindery) v1.40

Could this be a REAL versus INTEGER problem?





hideakii @ yahoo.com on 11/25/98 03:25:04 PM

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

To:   RATS-L @ efs.mq.edu.au
cc:   (bcc: Wai Lee)
Subject:  Exponentiation





Dear RATS Users

I got differencial results using 2 expression as follows:

1. x**(1.5)
2. x**(3/2)

and latter results was not one which I expected.
Was it incorrect expression??
I have checked Manual(1-29,30), but I could not find explanations
related to this.

Thanks in advance

ISHIGAMI Hideaki
Nippon Sports Science Univ.
_________________________________________________________
DO YOU YAHOO!?
Get your free @yahoo.com address at http://mail.yahoo.com








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

From: "Christopher F Baum" <baum@bc.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Exponentiation
Date: Wed, 25 Nov 1998 15:33:55 -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) [1.3.4, s/n P020-300786-009] (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

RATS treats expressions containing solely integers with integer division,
i.e. 3/2 == 1, 1/3 == 0. Always include decimals if you're using an
expression that involves a floating point power.

Kit Baum

--On Wed, Nov 25, 1998 12:25 -0800 hideakii <hideakii@yahoo.com> wrote: 

> Dear RATS Users
> 
> I got differencial results using 2 expression as follows:
> 
> 1. x**(1.5)
> 2. x**(3/2)
> 
> and latter results was not one which I expected.
> Was it incorrect expression??
> I have checked Manual(1-29,30), but I could not find explanations
> related to this.
> 
> Thanks in advance
> 
> ISHIGAMI Hideaki
> Nippon Sports Science Univ.
> _________________________________________________________
> DO YOU YAHOO!?
> Get your free @yahoo.com address at http://mail.yahoo.com
> 



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

From: Michael Hanson <mhanson@econ.lsa.umich.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Exponentiation
Date: Wed, 25 Nov 1998 15:32:23 -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) v1.40

	As I'm sure others will tell you, 3/2 = 1 because any operation
upon integers returns an integer.  (RATS math!)  But 3.0/2 = 3/2.0 =
3.0/2.0 = 1.5.  Somewhere in the manual is does explain that, although
given the number of times this question pops up on the mailing list,
perhaps the next version of the manual will place this information in
several prominent locations.

					-- Mike

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


On Wed, 25 Nov 1998, hideakii wrote:

> Dear RATS Users
> 
> I got differencial results using 2 expression as follows:
> 
> 1. x**(1.5)
> 2. x**(3/2)
> 
> and latter results was not one which I expected.
> Was it incorrect expression??
> I have checked Manual(1-29,30), but I could not find explanations
> related to this.
> 
> Thanks in advance
> 
> ISHIGAMI Hideaki
> Nippon Sports Science Univ.
> _________________________________________________________
> DO YOU YAHOO!?
> Get your free @yahoo.com address at http://mail.yahoo.com
> 
> 


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

From: hideakii <hideakii@yahoo.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: garch.src
Date: Wed, 25 Nov 1998 12:40:02 -0800 (PST)
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mercury MTS (Bindery) v1.40

Dear RATS Users

In garch.src, which is uploaded at Estima Web, there is procedure to
check serial correlation for normalized residuals and squared one,
using method developed by Ljung and Box(1979).

I think this test statistic has chi-squared distribution with degree
of freedom m, in which m is choiced lag order for checking serial
correlation.
In this procedure, however, degree of freedom is m-p, in which p is
lag order for autoregressive term of mean equation, not conditional
variance equation.

Is my understanding about L-B test incorrect? or is it that procedure
which is incorrect?  

Thanks in advance

ISHIGAMI Hideaki
Nippon Sports Science Univ.
_________________________________________________________
DO YOU YAHOO!?
Get your free @yahoo.com address at http://mail.yahoo.com


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

From: Nesmith_T <Nesmith_T@BLS.GOV>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: RE: Exponentiation
Date: Wed, 25 Nov 1998 15:49:41 -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>
Mime-Version: 1.0
X-Mailer: Internet Mail Service (5.5.1960.3) (via Mercury MTS (Bindery) v1.40)
Content-Type: text/plain

Dear ISHIGAMI

In the second expression you have integers rather than real numbers.  If
I remember correctly integer division truncates the result by dropping
the remainder, so the outcome of the second expression should be x.  The
proper expression is 

x**(3./2.)

The periods let RATS know the numbers are real and should result in the
same outcome as expression one, although the first expression is still
preferable.

Travis

> ----------
> From: 	hideakii[SMTP:hideakii@yahoo.com]
> Sent: 	Wednesday, November 25, 1998 3:25 PM
> To: 	RATS Discussion List
> Subject: 	Exponentiation
> 
> Dear RATS Users
> 
> I got differencial results using 2 expression as follows:
> 
> 1. x**(1.5)
> 2. x**(3/2)
> 
> and latter results was not one which I expected.
> Was it incorrect expression??
> I have checked Manual(1-29,30), but I could not find explanations
> related to this.
> 
> Thanks in advance
> 
> ISHIGAMI Hideaki
> Nippon Sports Science Univ.
> _________________________________________________________
> DO YOU YAHOO!?
> Get your free @yahoo.com address at http://mail.yahoo.com
> 

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

From: Nesmith_T <Nesmith_T@BLS.GOV>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: RE: Exponentiation
Date: Wed, 25 Nov 1998 15:52:38 -0500
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Mime-Version: 1.0
X-Mailer: Internet Mail Service (5.5.1960.3) (via Mercury MTS (Bindery) v1.40)
Content-Type: text/plain

Obviously, I had not gotten this message when I sent my reply.  One note
however is that at least in past versions of RATS I have had problems
when mixing reals and integers in expressions.  I know the manual says
that RATS will interpret the all numbers as reals in such an expression
but I find it best to always include a decimal point after each and
every real number.

Travis

> ----------
> From: 	Michael Hanson[SMTP:mhanson@econ.lsa.umich.edu]
> Sent: 	Wednesday, November 25, 1998 3:32 PM
> To: 	RATS Discussion List
> Subject: 	Re: Exponentiation
> 
> 	As I'm sure others will tell you, 3/2 = 1 because any operation
> upon integers returns an integer.  (RATS math!)  But 3.0/2 = 3/2.0 =
> 3.0/2.0 = 1.5.  Somewhere in the manual is does explain that, although
> given the number of times this question pops up on the mailing list,
> perhaps the next version of the manual will place this information in
> several prominent locations.
> 
> 					-- Mike
> 
> ======================================================================
> Michael S. Hanson                                    mhanson@umich.edu
> Department of Economics         http://www.econ.lsa.umich.edu/~mhanson
> University of Michigan
> 
> 
> On Wed, 25 Nov 1998, hideakii wrote:
> 
> > Dear RATS Users
> > 
> > I got differencial results using 2 expression as follows:
> > 
> > 1. x**(1.5)
> > 2. x**(3/2)
> > 
> > and latter results was not one which I expected.
> > Was it incorrect expression??
> > I have checked Manual(1-29,30), but I could not find explanations
> > related to this.
> > 
> > Thanks in advance
> > 
> > ISHIGAMI Hideaki
> > Nippon Sports Science Univ.
> > _________________________________________________________
> > DO YOU YAHOO!?
> > Get your free @yahoo.com address at http://mail.yahoo.com
> > 
> > 
> 

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

From: "Dr. Sean Michael Snaith" <ssnaith@uwimona.edu.jm>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Exponentiation
Date: Wed, 25 Nov 1998 15:57:43 -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: QUALCOMM Windows Eudora Pro Version 4.0 (via Mercury MTS (Bindery) v1.40)
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Turn the manual one more page to 1-31 - Integer vs. Real Numbers.

At 12:25 PM 11/25/98 -0800, you wrote:
>Dear RATS Users
>
>I got differencial results using 2 expression as follows:
>
>1. x**(1.5)
>2. x**(3/2)
>
>and latter results was not one which I expected.
>Was it incorrect expression??
>I have checked Manual(1-29,30), but I could not find explanations
>related to this.
>
>Thanks in advance
>
>ISHIGAMI Hideaki
>Nippon Sports Science Univ.
>_________________________________________________________
>DO YOU YAHOO!?
>Get your free @yahoo.com address at http://mail.yahoo.com
>=20




Dr. Sean Michael Snaith
Department of Economics
University of the West Indies
Kingston, 7
Jamaica
West Indies

TEL: (876) 977-1188=A0=20
FAX: (876) 977-1483


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

From: Sune Karlsson <Sune.Karlsson@hhs.se>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: garch.src
Date: Wed, 25 Nov 1998 22:22:49 +0100
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-version: 1.0
X-Mailer: Mozilla 4.5 [en] (WinNT; I) (via Mercury MTS (Bindery) v1.40)
Content-type: text/plain; charset=us-ascii
Content-transfer-encoding: 7BIT

hideakii wrote:
> 
> Dear RATS Users
> 
> In garch.src, which is uploaded at Estima Web, there is procedure to
> check serial correlation for normalized residuals and squared one,
> using method developed by Ljung and Box(1979).
> 
> I think this test statistic has chi-squared distribution with degree
> of freedom m, in which m is choiced lag order for checking serial
> correlation.
> In this procedure, however, degree of freedom is m-p, in which p is
> lag order for autoregressive term of mean equation, not conditional
> variance equation.
> 
> Is my understanding about L-B test incorrect? or is it that procedure
> which is incorrect?

Both. ;-) 
Both the procedure and you are correct, depending on how you view it.

The Ljung-Box statistic is asymptotically chi^2 with m degrees of
freedom. In small samples the lags of the dependent variable tends to
kill the first p residual autocorrelations and using a chi^2 with m-p
degrees of freedom gives a better approximation to the small sample
distribution of the statistic. I think there is a discussion of this in
the Ljung-Box paper,  or it may be in the earlier Box-Pierce paper.

/Sune
-- 
Sune Karlsson                      | stsk@hhs.se Fax: + 46 8 34 81 61
Stockholm School of Economics      | Phone: + 46 8 736 92 39
Box 6501, 113 83 Stockholm, Sweden | http://www.hhs.se/personal/SuneK/
http://www.hhs.se/stat/            | http://swopec.hhs.se/

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

From: "Christopher F Baum" <baum@bc.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: torats (no, not ohrats)
Date: Thu, 26 Nov 1998 08:20:58 -0500
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
X-Mailer: Mulberry (MacOS) [1.3.4, s/n P020-300786-009] (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Users of Stata sometimes find useful features lacking (especially when
torturing timeseries data). The admirable Stat/Transfer eases the pain of
relocating one's data to another package in many instances, but does not do
so for the RATS econometrics package (www.estima.com). Stat/Transfer can
make spreadsheets from .dta files, but the student trying to move her data
had a very large (1000+ variable) Stata dataset, and that might be
problematic... 

A simpler solution seemed to be to have Stata write the data, alone, via
outfile, and write the RATS program to read it (complete with all those
variable names). This is hindered considerably by Stata's lack of true
input/output facilities; i.e. in RATS, one can open a file, read arbitrary
contents (e.g. two integers and a real) from it, or write such to it. Stata,
from what I understand, cannot read from a file unless it thinks it's
reading a dataset, and cannot write to a file (except via outfile) other
than the logfile. So the logfile it must be...

The module TORATS.ADO, supplied with a filename, writes two files: the
currently active dataset (to filename.raw) and a logfile containing the RATS
code to read it (to filename.rat). Thanks to Nick Cox, one may specify -if-,
-in-, and an optional -varlist-, and string variables are automatically
omitted (RATS doesn't do strings). Missing data are the user's
responsibility. The module is available from IDEAS SSC at

http://ideas.uqam.ca/ideas/data/Softwares/bocbocodeS361501.html

I recognize that the clientele for this module may be small, but the
techniques used in getting Stata to write some identifying information to
facilitate the transfer of data to another application (your own FORTRAN or
C routine, other packages, etc.) may be far more general.

For those of an American persuasion, Happy Thanksgiving!

Kit Baum
SSC archive maintainer


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

From: Arturo Galindo <agalindo@students.uiuc.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Bartlett Window 
Date: Sat, 28 Nov 1998 15:11:49 -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: Mozilla 4.01 [en] (Win95; I) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Dear Everyone:

Does any one know how to use a Bartlett window for smoothing a
periodogram in RATS? I am not sure how to calculate a (modified)
Bartlett window as in Hamilton(1994) chapter 6 .
I am doing the following:

declare vector bartlett(freqs) **freqs is frequency size
do k = 0, freqs - 1
    compute bartlett(k) = 1 - ((k+1)/(1+freqs))
end do k

Am I doing it right?

In advance thanks,

Arturo Galindo
Department of Economics
University of Illinois at Urbana-Champaign


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

From: "Dr. Sean Michael Snaith" <ssnaith@uwimona.edu.jm>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Bartlett Window 
Date: Sat, 28 Nov 1998 16:50:52 -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: QUALCOMM Windows Eudora Pro Version 4.0 (via Mercury MTS (Bindery) v1.40)
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

I have not done the type of analysis you are referring to, but I *think*
there may be a problem with the do loop.  You have it running from 0 to
freqs -1, so you are trying to assign a value to the  zero-th element of
the bartlett vector.  Changing this zero to 1 allows the loop to run=
 properly.

I hope this helps. =20


At 03:11 PM 11/28/98 -0600, you wrote:
>Dear Everyone:
>
>Does any one know how to use a Bartlett window for smoothing a
>periodogram in RATS? I am not sure how to calculate a (modified)
>Bartlett window as in Hamilton(1994) chapter 6 .
>I am doing the following:
>
>declare vector bartlett(freqs) **freqs is frequency size
>do k =3D 0, freqs - 1
>    compute bartlett(k) =3D 1 - ((k+1)/(1+freqs))
>end do k
>
>Am I doing it right?
>
>In advance thanks,
>
>Arturo Galindo
>Department of Economics
>University of Illinois at Urbana-Champaign
>=20




Dr. Sean Michael Snaith
Department of Economics
University of the West Indies
Kingston, 7
Jamaica
West Indies

TEL: (876) 977-1188=A0=20
FAX: (876) 977-1483


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

From: "VIJAYA SARATHI N." <SARATHI@FPMLAN.iimahd.ernet.in>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: EM algorithm
Date: Sun, 29 Nov 1998 20:26:00 GMT+0530
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:  IIM, Ahmedabad
X-mailer: Pegasus Mail v3.31 (via Mercury MTS (Bindery) v1.40)

Dear colleagues,

I'm interested in executing the EM algorithm(Dempster, Laird, Rubin;
1977). Could you tell me if there is a RATS code for implementing 
this algorithm.

Thanks for your help.

Regards,

Sarathi
Indian Institute of Management 
India.

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

From: Simon.Van-Norden@hec.ca
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: EM algorithm
Date: Sun, 29 Nov 1998 10:15:42 -0500
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Organization: Ecole des Hautes Etudes Commerciales
X-Mailer: Mozilla 4.5 [en] (Win95; U) (via Mercury MTS (Bindery) v1.40)
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit

Dear Sarathi;

There is and there isn't.

The code you need for the E and the M step will differ depending on your
statistical model.

For example, in regime-switching models, the M step is just WLS (which can be
done in RATS by simply using LINREG with the SPREAD= option.)  However, the E
step is trickier (you need to find the smoothed probabilities of being in a
particular regime, for example using Kim's algorithm.)

Therefore;
1) There is no completely general RATS code for the EM algorithm,
2) There may be lots of code for EM algorithm for a specific class of models,
but you'd have to specify the kind of problem with which you are concerned.

Regards.

SvN

"VIJAYA SARATHI N." wrote:
> 
> Dear colleagues,
> 
> I'm interested in executing the EM algorithm(Dempster, Laird, Rubin;
> 1977). Could you tell me if there is a RATS code for implementing
> this algorithm.
> 
> Thanks for your help.
> 
> Regards,
> 
> Sarathi
> Indian Institute of Management
> India.

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

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

From: Rob Trevor <robt@efs.mq.edu.au>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: EM algorithm
Date: Mon, 30 Nov 1998 09:24:24 +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) v1.40

At 2:15 AM +1100 30/11/98, Simon.Van-Norden@hec.ca wrote:
>...
>2) There may be lots of code for EM algorithm for a specific class of models,
>but you'd have to specify the kind of problem with which you are concerned.
>...

Ieuan Morgan and I have a paper forthcoming in JBES on what we call a
R(ational)E(xpectations) algorithm for estimating GARCH models with
censored data. (Eg, price limits on a futures or stock exchange, missing
observations, etc.) If one assumes that the censoring impacts the data
generation process itself, our RE algorithm is an EM algorithm and hence
produces Maximum Likelihood estimates. Otherwise it produces approximate
Maximum Likelihood estimates. (We provide fairly extensive Monte Carlo
analysis of the approximation error as well as an empirical application.)

While we didn't use RATS for our paper, I intend to put some sample RATS
code together and release it as soon as publication is imminent. (Sorry, at
this stage I don't know when that will be. The sample code itself needs a
bit more work and testing and I probably won't be able to do that until the
new year.)

For anyone interested, the paper is available on the webin PDF format at
http://amethyst.efs.mq.edu.au/CMBF/paper18.pdf.

Cheers

Rob
-----------------------------------------------------------------------
Rob Trevor  Associate Professor
            Centre for Studies in Money, Banking and Finance
            Macquarie University, NSW, AUSTRALIA    2109

Internet:  robt@efs.mq.edu.au

VOICE: +61 (2) 9850-8447    FAX: +61 (2) 9850-7281 or +61 (2) 9418-1847
-----------------------------------------------------------------------

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