From: "Mr. Azali" <Mohammed@cardiff.ac.uk>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: VECM
Date: Tue, 6 May 1997 22:43:41 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
Content-transfer-encoding: 7BIT
X-mailer: Pegasus Mail for Windows (v2.33) (via Mercury MTS (Bindery) v1.30)

Hi,

Is there anyone who has a RATS procedure to estimate a VECM,
i.e. equation (8)  page 831 of King, Plosser, Stock & Watson, 
American Economic Review, Sept. 1991.

I would appreciate if you would contact me.

Regards,

M. Azali
Cardiff Business School
University of Wales Cardiff

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

From: "Li-gang Liu" <LLIU@iie.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Anticipated effect in Rational expectation models
Date: Thu, 8 May 1997 10:09:01 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>
Organization:  IIE
X-mailer: Pegasus Mail for Windows (v2.01) (via Mercury MTS (Bindery) v1.30)

Does anybody have a ready procedure in RATS that estimates the anticipated 
effect using the methodology proposed by Mishkin in his 1983 book  "A Rational 
Expectation Approach to Macroeconomics"?  

Thanks in advance!







------------------------------------------------------------
Li-gang Liu                             (O) 202-328-9000    |
Institute for International Economics   (H) 202-319-0768    |
11 Dupont Circle, Washington DC 20036   Fax 202-328-5432    |
------------------------------------------------------------

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

From: Economic Group <pgallagh@ozemail.com.au>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: kalman filter
Date: Fri, 9 May 1997 17:03:42 +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 (Bindery) v1.30

I'm using the Kalman filter to estimate a simple model with time varying
parameters.  As I understand it, the Kalman filter can be used in two ways.
The first is "prediction", where estimates of the time-varying parameters
at time t are produced based on the sample only up to time t.  The second
way is "smoothing", in which the path of the time varying parameter is
estimated at each point based on the whole sample.

In TSP, there is a simple way of choosing which algorithm to use, through
the use of the option SMOOTH.  The KALMAN command in RATS seems to be based
on the "prediction" method.  However, I want to smooth over the entire
sample (ie. estimate the evolution of the time varying parameters using
infomration from the entire sample).  My question thus is, how can I do
this in RATS?

James Vickery
Reserve Bank of Australia

Economic Analysis Department                           Email enquiries:
Economic Group                                         PH:  +612 9551 8817
Reserve Bank of Australia                              FAX: +612 9551 8833



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

From: Gregory Lypny <lypny@alcor.concordia.ca>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Multivariate t-distribution for GARCH
Date: Wed, 14 May 97 09:55:58 -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: Claris Emailer 1.1 (via Mercury MTS (Bindery) v1.30)
Mime-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"

Yes, I know it's ugly, but can anyone confirm whether this is the correct 
specification for the multivariate t distribution.

NU:-degrees of freedom
K=2 for bivariate case
H:-covariance matrix
U:-series of errors

..5*NU*LOG(NU)+%LNGAMMA(.5*(NU+K))-%LNGAMMA(.5*NU) $
-.5*LOG(%DET(((NU-2)/NU)*H))-.5*(NU+K)*LOG(NU+%QFORM(INV(((NU-2)/NU)*H),U))


Thanks,



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@ALCOR.CONCORDIA.CA
Montreal, QC H3G 1M8


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

From: dwatt@bank-banque-canada.ca (David Watt)
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Multivariate t-distribution for GARCH
Date: Wed, 14 May 1997 09:58:32 -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 (Bindery) v1.30



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

From: dwatt@bank-banque-canada.ca (David Watt)
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: Multivariate t-distribution for GARCH
Date: Wed, 14 May 1997 10:05:32 -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 (Bindery) v1.30

Sorry about the previous post....

> Yes, I know it's ugly, but can anyone confirm whether this is the correct 
> specification for the multivariate t distribution.
> 
> NU:-degrees of freedom
> K=2 for bivariate case
> H:-covariance matrix
> U:-series of errors
> 
> .5*NU*LOG(NU)+%LNGAMMA(.5*(NU+K))-%LNGAMMA(.5*NU) $
> -.5*LOG(%DET(((NU-2)/NU)*H))-.5*(NU+K)*LOG(NU+%QFORM(INV(((NU-2)/NU)*H),U))
> 

What I typically use is very close and is given by...

%LNGAMMA(.5*(NU+K)) - %LNGAMMA(.5*NU) - (.5*NU)*LOG(NU-2) - .5*LOG(%DET(H)) $
- (.5*K)*LOG(%PI) - .5*(NU+K)*LOG(1+((1/(NU-2))*%QFORM(INV(H),U)))


Dave Watt
Bank of Canada. 

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

From: Wai Lee <lee_wai@jpmorgan.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: estimate
Date: 20 May 97 15:23:36 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
X-Mailer: Mercury MTS (Bindery) v1.30

I want to store the coeffs from ESTIMATE, and I do

DECLARE VECT[SERIES] COEFF(NEQN)

I get the coeff in different series, but of course, the coefficients are in the 
first few rows and with all zeros in other entries of the time series.  Is 
there a better way to do this?  I tried RECT but it doesn't work.

Thanks.

W. Lee


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

From: "Christopher F. Baum" <BAUM@bcvms.bc.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: estimate
Date: Tue, 20 May 1997 19:42:39 -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

Wai Lee asked how to get the coefficients from a VAR system into a matrix:

comp neqn=3
comp nlag=4
comp ncoef=neqn*nlag+1
cal 1946 1 12
all neqn 1987:1
open data intrates.dat
data(org=var) 1948:1 1987:1 prime
data(org=var) 1964:1 1987:1 tb3mo
data(org=var) 1947:1 1987:1 tb10yr
system 1 to neqn
variables prime tb3mo tb10yr
lags 1 to nlag
det constant
end(system)
estimate / * 1
dec rect coef(ncoef,neqn)
make coef 1 ncoef
# 1 to neqn
write coef
end

Kit Baum
Boston College Econ

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

From: Christopher F Baum <baum@bc.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Review of RATS numerical accuracy
Date: Wed, 21 May 1997 10:16:01 -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: Simeon for Macintosh Version 4.1 Build (3) (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; CHARSET=US-ASCII

An interesting review of RATS--focusing on its numerical accuracy--has 
appeared in Journal of Applied Econometrics, 12:2, March-April 1997, 
pp. 181-190: B.D. McCullough, "A Review of RATS v4.2: Benchmarking 
Numerical Accuracy." He points out that RATS (like many other commonly 
used packages) is at times none too accurate with poorly-conditioned 
data, and suggests that more information be available in the 
documentation regarding the precise algorithms used for computation.

Kit Baum
---------------------------------------------------------------------
Christopher F Baum    Boston College Econ, Chestnut Hill MA 02167 USA         
baum@bc.edu  fax 617 552 2308  http://fmwww.bc.edu/EC-V/baum.fac.html



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

From: LYPNY@ALCOR.CONCORDIA.CA
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Maximizing original series vs. minimizing errors
Date: 27 May 97 20:16:36 -0400
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
X-Mailer: Cyberdog/2.0 (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="Cyberdog-MixedBoundary-0001DC16"
Content-Transfer-Encoding: 7bit


--Cyberdog-MixedBoundary-0001DC16
Content-Type: image/jpeg
Content-Transfer-Encoding: base64
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--Cyberdog-MixedBoundary-0001DC16
X-Fontfamily: New%20York
X-Fontsize: 9
Content-Type: text/enriched; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<SMALLER><SMALLER>


</SMALLER></SMALLER><X-FONTSIZE><PARAM>12</PARAM>Hi,


This one's for Rob Trevor, or anyone interested in GARCH estimations.


I've been running multivariate GARCH models on my Mac by maximizing
the negative of the log-likelihood of the residuals.  I've been using
code very similar to Rob Trevors, and, in some cases identical.


However, minimizing the log-likelihood of the residuals is not
necessarily the same as maximizing the log-likelihood of their
underlying series.  How can I re-specify my mean equations, so that I
can do the latter.


For example, if a data series is called y, Rob Trevor's code might
look like this:


FRML RESID =3D Y - B0,


where RESID would be fed into an error series "u".


Would I simply add  the additional FRML 


FRML NewY =3D B1 + RESID


and feed NewY into the maximization?



Regards,


Greg Lypny


Concordia University

Montreal

</X-FONTSIZE><X-FONTSIZE><PARAM>12</PARAM><FONTFAMILY><PARAM>Palatino</=
PARAM>

	</FONTFAMILY></X-FONTSIZE>
--Cyberdog-MixedBoundary-0001DC16--


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