From: jacques tebeka <101677.76@compuserve.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: roots of an AR(p) model
Date: Fri, 10 Jan 1997 10:23:36 -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>
X-Mailer: Mercury MTS (Bindery) v1.30
Can someone help me to solve the characteristics equation of an AR(p)
model. I do not see how to proceed.
For example, i have an AR(3) model: y = c + a1*y{1} + a2*y{2} + a3*y{3} and
i would like to find z such that : 1 - a1*z - a2*z**2 - a3*z**3 = 0.
Thanks for your help.
Jacques Tebeka
e-mail: 101677.76@compuserve.com
---------- End of message ----------
From: "Christopher F. Baum" <BAUM@bcvms.bc.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Jacques'
Date: Fri, 10 Jan 1997 16:40:41 -0400 (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 (Bindery) v1.30
* calculate stability of an AR(3) model
* place the estimated coefficients in the first row of mtx
* remaining rows are [ I | 0 ] where I is of dimension (p-1)
*
* For stability, all moduli must be less than unity
*
* CFBaum [baum@bc.edu]
*
dec rect mtx(3,3)
input mtx
1.1 -0.5 0.3
1.0 0.0 0.0
0.0 1.0 0.0
write mtx
eigen(cvalues=eigval) mtx
write eigval
dec vec rpt(3) img(3) modulus(3)
comp stab=0.0
do i=1,3
compute rpt(i)=%real(eigval(i))
compute img(i)=%imag(eigval(i))
compute modulus(i)=sqrt(rpt(i)**2+img(i)**2)
if modulus(i)>stab
compute stab=modulus(i)
enddo
write modulus
if stab<1
write "system is stable: largest root is " stab
else
write "system is unstable: largest root is " stab
endif
end
---------- End of message ----------
From: colletaz <colletaz@univ-orleans.fr>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: roots of an AR(p) model
Date: Sun, 12 Jan 1997 15:21:35 +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 3.0 (Win95; I) (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
jacques tebeka wrote:
>
> Can someone help me to solve the characteristics equation of an AR(p)
> model. I do not see how to proceed.
> For example, i have an AR(3) model: y = c + a1*y{1} + a2*y{2} + a3*y{3} and
> i would like to find z such that : 1 - a1*z - a2*z**2 - a3*z**3 = 0.
> Thanks for your help.
> Jacques Tebeka
> e-mail: 101677.76@compuserve.com
Do you have access to a program like Mathematica or something like that
(of course, it seems like using a very big hammer to solve only such
problem ... but it's so easy)
---------- End of message ----------
From: "Christopher F. Baum" <BAUM@bcvms.bc.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: roots of an AR(p) model
Date: Mon, 13 Jan 1997 21:24:21 -0400 (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 (Bindery) v1.30
As my posting last Friday indicated, RATS is perfectly capable of solving
this problem, since it can calculate the eigenvalues of a nonsymmetric
matrix (in general complex). I used Mathematica to check my code, but
it surely is not needed in this case (and it is very useful to be able to
analyse the stability of an estimated dynamic model within the RATS
environment in which you estimated it!)
Kit Baum
baum@bc.edu
---------- End of message ----------
From: jacques tebeka <101677.76@compuserve.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: nonlinear test
Date: Wed, 15 Jan 1997 04:55:55 -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: Mercury MTS (Bindery) v1.30
In most nonlinear test (LM test, Cf. Granger & Terasvirta), we need to
calculate something like that:
from that vector X = (X1,X2,...Xk), we need X(t)*Z(t)**j or something like
that. Does anyone have a solution to do that automatically (of course, i
can calculate each series manually). I think the solution is to define a
vector[series] but after how to introduce all the new variables in a
regression. I tried to introduce the new vector of series in a regression
but rats 'reject' that.
Thanks for your help concerning that question.
Does anyone have a program to test nonlinearity using Tsay (1989)
procedure.
I saw 2 or 3 months ago, that someone was looking for the code of Hamilton
procedure (process with change in regime), i do not know if someone answer
him because i am interesting too.
Thanks for your help to all that (interresting?) questions.
Tebeka.
---------- End of message ----------
From: Gregory Lypny <lypny@alcor.concordia.ca>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: RATS Discussion List
Date: Wed, 15 Jan 97 08:12:56 -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 (Bindery) v1.30)
Mime-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
UNSUBSCRIBE
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: Kim Minho <kimmh@moak.chonbuk.ac.kr>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Q) Default Directory
Date: Tue, 24 Jan 1995 22:44:23 +0900
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
X-Mailer: Mozilla 3.01 (Win95; I) (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Is there any way I can set the default directory permanently in the
directory menu of the file option?
---------- End of message ----------
From: "Ulrich Leuchtmann" <ULEUCHTMANN@discordia.wiwi.uni-bielefeld.de>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Q) Default Directory
Date: Fri, 24 Jan 1997 15:15:04 GMT+0100
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Organization: Fak. WIWI Uni Bielefeld
X-Mailer: Pegasus Mail for Windows (v2.23DE) (via Mercury MTS (Bindery) v1.30)
Hi Kim,
in a message from Tue, 24 Jan 1995 22:44:23 +0900 you wrote:
> Is there any way I can set the default directory permanently in the
> directory menu of the file option?
If you're using the Windows-Version of RATS V4.*, you can get the
effect you want by selecting the "Working Directory" settings from
the "Program Properties" menu of Windows (I'm using the German
Windows version, so I'm not sure about the exact English names).
Pop up that menu by marking the RATS program icon in your
program manager and hitting Alt+Return. Then you can edit the
"Working directory" entry and choose the directory, which RATS uses
as the permanent default directory.
Best regards
Uli
=============================================================
Ulrich Leuchtmann
University of Bielefeld
Faculty of Economics
P.O.Box 10 01 31 Tel.: (+49)(0) 521-106-4870
33501 Bielefeld Fax: (+49)(0) 521-106-2994
Germany email: ULeucht@wiwi.uni-bielefeld.de
=============================================================
---------- End of message ----------
From: "Estima" <estima@estima.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: nonlinear test
Date: Fri, 24 Jan 1997 10:08:13 -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 (v2.42) (via Mercury MTS (Bindery) v1.30)
> In most nonlinear test (LM test, Cf. Granger & Terasvirta), we need to
> calculate something like that:
> from that vector X = (X1,X2,...Xk), we need X(t)*Z(t)**j or something like
> that. Does anyone have a solution to do that automatically (of course, i
> can calculate each series manually). I think the solution is to define a
> vector[series] but after how to introduce all the new variables in a
> regression. I tried to introduce the new vector of series in a regression
> but rats 'reject' that.
In general, you shouldn't have any problem using a VECTOR of SERIES
in a regression. For example:
DEC VEC[SER] VS(3)
* Set the series as appropriate:
SET VS(1) = ???
SET VS(2) = ???
SET VS(3) = ???
You can now refer either to one particular series in the vector:
LINREG Y
# X Z VS(1)
or you can include all the series in the VECTOR by just specifying
the VECTOR name:
LINREG Y
# X Z VS
Let me know if you still have problems with this.
Sincerely,
Tom Maycock
Estima
--
+-----------------------------+-----------------------------------------+
| Estima | |
| P.O. Box 1818 | Voice: (847) 864-8772 |
| Evanston, IL 60204-1818 | Fax: (847) 864-6221 |
| U.S.A | BBS: (847) 864-8816 |
| e-mail: estima@estima.com | CompuServe: 73140,2202 |
|-----------------------------------------------------------------------|
| Web Site: http://www.estima.com |
| RATS Internet Mailing List: New members can join by sending e-mail to |
| MAISER@EFS.MQ.EDU.AU with the message: SUBSCRIBE RATS-L |
+-----------------------------------------------------------------------+
---------- End of message ----------
From: "Estima" <estima@estima.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Q) Default Directory
Date: Fri, 24 Jan 1997 10:08:13 -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 (v2.42) (via Mercury MTS (Bindery) v1.30)
>
> in a message from Tue, 24 Jan 1995 22:44:23 +0900 you wrote:
> > Is there any way I can set the default directory permanently in the
> > directory menu of the file option?
>
> If you're using the Windows-Version of RATS V4.*, you can get the
> effect you want by selecting the "Working Directory" settings from
> the "Program Properties" menu of Windows (I'm using the German
> Windows version, so I'm not sure about the exact English names).
>
> Pop up that menu by marking the RATS program icon in your
> program manager and hitting Alt+Return. Then you can edit the
> "Working directory" entry and choose the directory, which RATS uses
> as the permanent default directory.
>
This is correct. Note that in using Windows 95 (or NT 4.0) the field
is entitled "Start In" rather than "Working Directory." You can get
to this field by hitting <Alt>+<Enter> or by clicking on the icon
with the right mouse button.
If you are using the DOS version of RATS, you can use the
"Configure" operation on the File menu. The third screen lets you
set a default start-up path. Note that if you specify a path here,
it will be used by both DOS and Windows versions of RATS, but the
only way to change it is via the DOS version.
Sincerely,
Tom Maycock
Estima
--
+-----------------------------+-----------------------------------------+
| Estima | |
| P.O. Box 1818 | Voice: (847) 864-8772 |
| Evanston, IL 60204-1818 | Fax: (847) 864-6221 |
| U.S.A | BBS: (847) 864-8816 |
| e-mail: estima@estima.com | CompuServe: 73140,2202 |
|-----------------------------------------------------------------------|
| Web Site: http://www.estima.com |
| RATS Internet Mailing List: New members can join by sending e-mail to |
| MAISER@EFS.MQ.EDU.AU with the message: SUBSCRIBE RATS-L |
+-----------------------------------------------------------------------+
---------- End of message ----------
From: <Bao.Wang@DAL.frb.org>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Float Precision.
Date: Tue, 28 Jan 1997 11:02:42 +1100
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
X-Mailer: Mercury MTS (Bindery) v1.30
Hi, Friends:
Does anyone know what precision the float number in matrix of RATS has ?
(single precision float number or double precision). It is so important that
when you have to use the matrix inverse operation in some tests.
I want to ask this question because I got different results with RATS and
other software when I did a Monte Carlo experimentation. One friend of mine
suspect that RATS implements in single precision and causes this difference.
I am looking forward to the reply. Thanks
Baoyuan Wang
Dept of Econ
SMU, Dallas
---------- End of message ----------
From: "Estima" <estima@estima.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Float Precision.
Date: Tue, 28 Jan 1997 10:00:03 -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 (v2.42) (via Mercury MTS (Bindery) v1.30)
>
> Does anyone know what precision the float number in matrix of RATS has ?
> (single precision float number or double precision). It is so important that
> when you have to use the matrix inverse operation in some tests.
>
> I want to ask this question because I got different results with RATS and
> other software when I did a Monte Carlo experimentation. One friend of mine
> suspect that RATS implements in single precision and causes this difference.
>
Dear Mr. Wang:
Nope, RATS uses double precision, not single precision. If you can't
resolve the problems you're running into, let me know.
Sincerely,
Tom Maycock
Estima
--
+-----------------------------+-----------------------------------------+
| Estima | |
| P.O. Box 1818 | Voice: (847) 864-8772 |
| Evanston, IL 60204-1818 | Fax: (847) 864-6221 |
| U.S.A | BBS: (847) 864-8816 |
| e-mail: estima@estima.com | CompuServe: 73140,2202 |
|-----------------------------------------------------------------------|
| Web Site: http://www.estima.com |
| RATS Internet Mailing List: New members can join by sending e-mail to |
| MAISER@EFS.MQ.EDU.AU with the message: SUBSCRIBE RATS-L |
+-----------------------------------------------------------------------+
---------- End of message ----------
From: slk44@cc.usu.edu
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Float Precision.
Date: Tue, 28 Jan 1997 11:42:34 -0600 (MDT)
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: Mercury MTS (Bindery) v1.30
unsubscribe
---------- End of message ----------
From: rs <rjschoen@earthlink.net>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Float Precision.
Date: Tue, 28 Jan 1997 15:55:06 -0800
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
X-Mailer: Mozilla 3.0Gold (Win16; I) (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
unsubscribe
---------- End of message ----------
From: Rob Trevor <robt@efs.mq.edu.au>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: ADMIN Note
Date: Thu, 30 Jan 1997 10:06:30 +1100
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
X-Mailer: Mercury MTS (Bindery) v1.30
Dear Colleagues
We now have 265 subscribers to this list - so PLEASE be careful about
sending 'unsubscribe' and other admin messages to the list and wasting the
valuable time of 264 other people!
If you so wish, you may unsubscribe at any time by sending an e-mail message to
MAISER@EFS.MQ.EDU.AU
with any subject in the subject line and containing the one line message
UNSUBSCRIBE RATS-L
DO NOT send such admin messages to the list address RATS-L@efs.mq.edu.au.
Finally, let me briefly state my policy on bounced mail. I receive about
20 bounced mail messages for each message that goes out on this list. Some
of these are due to people's email disk quota being filled, some are due to
people subscribing from email accounts on computers which are inaccessible
for long periods (my mailer tries for 5 days), some are due to people
subscribing from corrupted email accounts (ie the address they subscribe
from can not be reached from outside). From time to time I endeavor to
contact people who (or their postmaster) who have addresses which regularly
bounce. However, I have (for obvious reasons) often found it impossible to
contact such people.
If you have a colleague who thinks that they are subscribed to RATS-L but
does NOT get THIS message, then please ask them to drop me a line with a
fax number where they can be contacted.
Many thanks
Rob Trevor
robt@efs.mq.edu.au
---------- End of message ----------