From: William Veloce <wmveloce@spartan.ac.BrockU.CA>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: restricting boxjenk parameters
Date: Wed, 3 Jul 1996 13:31:59 +0059 (EDT)
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 v1.21


Does anyone know whether it is possible to restrict parameters
in ARIMA, transfer function models etc. in RATS ?
I tried the restrict command it only can be applied to linreg.

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

From: mdurand@orstom.fr (Marie-H\il\hne Durand)
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: seasonal unit root tests
Date: Wed, 10 Jul 1996 19:28:37 +0100
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
X-Mailer: Windows Eudora Version 1.4.4 (via Mercury MTS v1.21)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"

Dear Rats users,

I am looking for people having coded programms for seasonal unit root test
as established by Canova and Hansen (Are seasonal patterns constant over
time,  a test for seasonal stability, Journal of Business and Economics
Statistics,july 1995, vol.13,n.3)

Thank you


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

From: Rob Trevor <robt@efs.mq.edu.au>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: RATS-L Archives
Date: Fri, 12 Jul 1996 10:28:14 +1000
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 v1.21

Dear Marie-H=E9l=E8ne and others affected,

>I have subsribed to the rats-l list and I try to get the list of file us=
ing
>the command INDEX, but I always recieve a delivery failure notification
>saying that the adress MAISER@EFS.MQ.EDU.AU is not found or locked. How =
can
>I get out of this problem?
>Thank you very much
>Marie-H=E9l=E8ne Durand

The sys admin has identified the cause of this problem, and is planning t=
o
rectify it on Sunday evening (local time) when the server hardware will
have its disk space significantly upgraded. So please wait until Monday
your local times (the sun rises first down here - well almost) before
trying to retrieve any of the archives.

Cheers

Rob Trevor
RATS-L-OWNER@EFS.MQ.EDU.AU




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

From: Rob Trevor <robt@efs.mq.edu.au>
To: RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Archives back up
Date: Wed, 17 Jul 1996 12:23:59 +1000
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 v1.21

Just a quick note to say that the RATS-L archives are again functioning as
advertised. More importantly, the server's disk space has been inreased
dramatically so that the rash of problems we've had since late April should
now be a thing of the past. (At least that's what I've been told by the sys
admin!)

Cheers

Rob Trevor
RATS-L-OWNER@EFS.MQ.EDU.AU



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

From: THIERRY.MICHEL@prevision.finances.fr
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: simulations with CATS
Date: Wed, 17 Jul 1996 13:16:10 +0200
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: quoted-printable
X-Mailer: Mercury MTS v1.21

     Dear RATS users

          I'm estimating a simple cointegration model with no exogenous
          variables using CATS, but I can't find an easy way to get the
          estimated results back into my main RATS program.
          I'd like to do some forecasting, that is generate dynamics for =
the
          whole complete system , using the long run relations (found in =
the
          %Pimat matrix in CATS) and short run dynamics (found ??) .
          What is the best way to do that ? Is there a trick to actually =
use the
          estimated model for forecasting (maybe with the recursive optio=
n) ?

          Many thanks..
                     Thierry.


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

From: Sune Karlsson <stsk@HHS.SE>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: simulations with CATS
Date: Wed, 17 Jul 1996 14:35:24 +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: Stockholm School of Economics
MIME-version: 1.0
X-Mailer: Mozilla 2.02 (WinNT; I) (via Mercury MTS v1.21)
Content-type: text/plain; charset=us-ascii
Content-transfer-encoding: 7bit

THIERRY.MICHEL@prevision.finances.fr wrote:
>
>      Dear RATS users
>
>           I'm estimating a simple cointegration model with no exogenous
>           variables using CATS, but I can't find an easy way to get the
>           estimated results back into my main RATS program.
>           I'd like to do some forecasting, that is generate dynamics for the
>           whole complete system , using the long run relations (found in the
>           %Pimat matrix in CATS) and short run dynamics (found ??) .
>           What is the best way to do that ? Is there a trick to actually use the
>           estimated model for forecasting (maybe with the recursive option) ?

I have a Rats procedure which does forecasts from "simple" cointegration models.
No restriction on the adjustment coefficients (alpha or gamma depending on your
choice of notation) and only the restriction beta =
--
----------------------------------------------------------------------
Sune Karlsson                                 Internet:    STSK@HHS.SE
Stockholm School of Economics                 Phone:  + 46 8 736 92 39
Box 6501, 113 83 Stockholm, Sweden            Fax:     + 46 8 34 81 61

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

From: Sune Karlsson <stsk@HHS.SE>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: simulations with CATS
Date: Wed, 17 Jul 1996 14:42:13 +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>
Organization: Stockholm School of Economics
MIME-version: 1.0
X-Mailer: Mozilla 2.02 (WinNT; I) (via Mercury MTS v1.21)
Content-type: text/plain; charset=us-ascii
Content-transfer-encoding: 7bit

Ooops, must have pressed the wrong button. How embarrassingÉ

Here is what I intended to say,

I have a Rats procedure which does forecasts from "simple" cointegration
models. No restriction on the adjustment coefficients (alpha) and only the
restriction beta = H*phi on the cointegrating vector(s).

I can send it to you if you want, but be warned that some of the other
things it does are probably wrong. In particular I am afraid that the
impulse responses and variance decompositions doesn't account for the
cointegration in a proper way.

/Sune
--
----------------------------------------------------------------------
Sune Karlsson                                 Internet:    STSK@HHS.SE
Stockholm School of Economics                 Phone:  + 46 8 736 92 39
Box 6501, 113 83 Stockholm, Sweden            Fax:     + 46 8 34 81 61

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

From: mdurand@orstom.fr (Marie-H\il\hne Durand)
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: seasonal unit root tests
Date: Wed, 17 Jul 1996 22:11:46 +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 Version 1.4.4 (via Mercury MTS v1.21)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"

Dear Rats users,

I am looking for people having coded programms for seasonal unit root test
as established by Canova and Hansen (Are seasonal patterns constant over
time,  a test for seasonal stability, Journal of Business and Economics
Statistics,july 1995, vol.13,n.3)

Thank you


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

From: Hameed Abid <iw94@jove.acs.unt.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Rats for Windows
Date: Fri, 19 Jul 1996 12:08:06 -0500 (CDT)
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 v1.21

Hello folks,
   I'm thinking of buying a copy of WinRats and thought would be nice to
get feedback from someone who has used it . Is it easy to get impulse
reponse graphs relative to dos versions.
Thanks,
Abid Hameed

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

From: Gregory Lypny <lypny@alcor.concordia.ca>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Rats for Windows
Date: Fri, 19 Jul 96 14:36:31 -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: Claris Emailer 1.1 (via Mercury MTS v1.21)
Mime-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"

I use MacRats and find it both powerful and easy to use.  I recently
purchased TSP 4.3 for the PowerMac, and it does not appear to be nearly
as interactive as Rats.  I don;t know about impulse response graphs.



Gregory J. Lypny
Associate Professor of Finance
Director, Centre for Instructional Technology

Mailing address:                            Contact:
-------------------------------------------------------------------
Finance Department                          514.848.2926 (voice)
Concordia University                        514.848.4500 (fax)
1455 De Maisonneuve Boulevard West          LYPNY@VAX2.CONCORDIA.CA
Montreal, QC H3G 1M8


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

From: CynthiaMcN@aol.com
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Rats for Windows
Date: Fri, 19 Jul 1996 16:01:41 -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>
X-Mailer: Mercury MTS v1.21

WinRats is wonderful?  The impulse graphs are simple . . . just pull up the
"impulse.prg" and make appropriate substitutions.  It's great and well worth
the investment . . . no comparison with Dos.

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

From: CynthiaMcN@aol.com
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Rats for Windows
Date: Fri, 19 Jul 1996 16:03:37 -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>
X-Mailer: Mercury MTS v1.21

Oops!! the first was supposed to read "WinRats is wonderful !!"

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

From: David Aadland <aadland@darkwing.uoregon.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: SVAR Procedure
Date: Fri, 19 Jul 1996 15:18:38 -0700
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 v1.21)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"

Has anyone used the SVAR procedure for imposing Blanchard and Quah's
long-run restriction in a bivariate VAR?  If so, how do you recommend
getting starting values for gamma (i.e., the free elements in C)?

David Aadland
Ph.D. Candidate
Dept. of Economics
University of Oregon


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

From: David Montgomery <dmontgom@hawaii.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject:
Date: Wed, 24 Jul 1996 20:01:29 -1000
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 v1.21)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"

Dear Rats Users,

        I am looking for the source code named LNGAMMA.src.  This code
calculates the log of the gamma function.  I believe this code came with
rats but for some reason, I don't have it.  If someone has it, can you
please email it to me ASAP.

        For those interested in MLE of a fractionally differenced ARFIMA
model.  I have a program that will calculate up to an ARFIMA(0,d,3) on my
homepage at www2.hawaii.edu/~dmontgom.

Thanks

David Montgomery


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

From: Rob Trevor <robt@efs.mq.edu.au>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: LNGAMMA
Date: Thu, 25 Jul 1996 16:20:55 +1000
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 v1.21

David

I think its build into RATS 4.20, not as an external procedure.

Rob

At 4:01 PM 25/7/96, David Montgomery wrote:
>É
>        I am looking for the source code named LNGAMMA.src.  This code
>calculates the log of the gamma function.  I believe this code came with
>rats but for some reason, I don't have it.  If someone has it, can you
>please email it to me ASAP.
>É



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

From: David Montgomery <dmontgom@hawaii.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject:
Date: Wed, 24 Jul 1996 21:57:45 -1000
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 v1.21)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"

Never mind.  I found out that %LNGAMMA is a built in function.  I thought it
was an external procedure.  Thanks


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

From: David Montgomery <dmontgom@hawaii.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject:
Date: Wed, 24 Jul 1996 23:57:38 -1000
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 v1.21)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"

Dear Rats Users,

        The %lngamma defined in rats does not help at all and this function
will not work for negative values.  I need to calculate the gamma function
for large negative values from -170 to -500.  If anyone has a program that
does this or if anyone is an expert on the gamma function then please
contact because I need the help!!

Thanks

David Montgomery


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

From: dprieul@lehman.com (David Prieul)
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: RE: %Lngamma
Date: Thu, 25 Jul 96 12:59:29 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>
X-Mailer: Mercury MTS v1.21

% Dear Rats Users,
%
%        The %lngamma defined in rats does not help at all and this function
% will not work for negative values.  I need to calculate the gamma function
% for large negative values from -170 to -500.  If anyone has a program that
% does this or if anyone is an expert on the gamma function then please
% contact because I need the help!!

I was aware of a problem with %Lngamma working with scalars but not vectors. I do not know what you want to perform but the way I used to solve the problem working with univariate models was to use a approximation of the function (as   available in Abram

You can otherwise try to adapt some Fortran code in Rats, I have a program that does that.

David Prieul
dprieul@lehman.com

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

From: David Montgomery <dmontgom@hawaii.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: RE: %Lngamma
Date: Thu, 25 Jul 1996 03:03:58 -1000
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 v1.21)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"

Dear David,

        If you have a code that will approximate the gamma function in Rats,
I would be gald to get a copy.  I am using this code for MLE of a
fractionally differenced ARFIMA model.

Thanks for your help

David Montgomery


At 01:59 AM 7/25/96 -1000, you wrote:
>% Dear Rats Users,
>%
>%        The %lngamma defined in rats does not help at all and this function
>% will not work for negative values.  I need to calculate the gamma function
>% for large negative values from -170 to -500.  If anyone has a program that
>% does this or if anyone is an expert on the gamma function then please
>% contact because I need the help!!
>
>I was aware of a problem with %Lngamma working with scalars but not
vectors. I do not know what you want to perform but the way I used to solve
the problem working with univariate models was to use a approximation of the
function (as available in Abram
>
>You can otherwise try to adapt some Fortran code in Rats, I have a program
that does that.
>
>David Prieul
>dprieul@lehman.com
>


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

From: XIAO <yxiao@falcon.cc.ukans.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject:
Date: Fri, 26 Jul 1996 17:17:25 -0500 (CDT)
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 v1.21


Dear Rats Users:

I want to use a loop to generate a series at each step, how can I save
these series? I thought about declaring a matrix ahead and save one
series as one columm (using the step index) of that matrix, but I don't
know how to make it work. If anyone know how to do it, would you please
help me?

Thank you very much!



Michelle Xiao

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

From: Norman Morin <nmorin@weber.ucsd.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re:
Date: Fri, 26 Jul 1996 16:17:48 -0700
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 v1.21

At 5:17 PM -0500 on 7/26/96, XIAO wrote:
>Dear Rats Users:
>
>I want to use a loop to generate a series at each step, how can I save
>these series? I thought about declaring a matrix ahead and save one
>series as one columm (using the step index) of that matrix, but I don't
>know how to make it work. If anyone know how to do it, would you please
>help me?


You might try a vector of series.

Call it seriesvec; so somewhere you'll have to have:

DECLARE VECT[series] seriesvec

Suppose it is going to be constist of n series, some at some point
you'll have

DIMENSION seriesvec(n)

Then in your loop you'll do whatever you need to do to
generate the series, e.g. creating n series of AR(1)'s,

DO ii = 1,n
   SET(first=0.) SERIESVEC(ii) start end = .75*SERIESVEC(ii){1} + %ran(1.)
END DO ii

   or

DO ii = 1,n
  SET SERIESVEC(ii) start end = 0.
  SET e start end = %ran(1.)
  DO jj = start+1,end
     COMPUTE SERIESVEC(ii)(jj) = .75*SERIESVEC(ii)(jj-1) + e(jj)
  END DO jj
END DO ii

I hope this helps.

--Norm


                       Norman J. Morin
                     nmorin@weber.ucsd.edu
-------------------------------------------------------------
Department of Economics * University of California, San Diego
        9500 Gilman Drive * La Jolla, CA  92023-0508
-------------------------------------------------------------



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