From: Petri Kangas <petkang@utu.fi>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: More on multivariate GARCH-M
Date: Mon, 04 Aug 1997 08:24:49 +0300
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 turku
X-Mailer: Mozilla 4.01 (Macintosh; I; PPC) (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1; x-mac-type="54455854"; x-mac-creator="4D4F5353"
Content-Transfer-Encoding: QUOTED-PRINTABLE

Hi!

I have tried to estimate time-varying risk premium in the term structur=
e
of interest rates in CAPM framework. I keep getting errormessages (the
fatal ones that crash your program) like "floating point error: square
root of negative number". I have tried bivariate GARCH-M using BEKK
parametrization but it does not seem to work. What=B4s wrong with my
program. i think the problem is in my program, not with RATS.

I attach my version of the program with this message

best wishes Petri Kangas

ps. the (original) program was given me by Mr Philippe Protin. Thank Yo=
u
for that!

--
Petri Kangas
Assistant
Department of Economics
University of Turku
FIN-20014  TURKU  FINLAND
e-mail: petkang@utu.fi
tel: +358-2-333 5884
fax: +358-2-333 5893
=20


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

From: Petri Kangas <petkang@utu.fi>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: oops! I forgot the program
Date: Mon, 04 Aug 1997 08:27:23 +0300
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 turku
X-Mailer: Mozilla 4.01 (Macintosh; I; PPC) (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="------------89026F74002D06D097E66A7A"

This is a multi-part message in MIME format.
--------------89026F74002D06D097E66A7A
Content-Type: text/plain; charset=us-ascii; x-mac-type="54455854"; x-mac-creator="4D4F5353"
Content-Transfer-Encoding: 7bit

Hi heres the program (multivariate GARCH-M with BEKK parametrization)

--
Petri Kangas
Assistant
Department of Economics
University of Turku
FIN-20014  TURKU  FINLAND
e-mail: petkang@utu.fi
tel: +358-2-333 5884
fax: +358-2-333 5893
 

--------------89026F74002D06D097E66A7A
Content-Type: text/plain; charset=us-ascii; x-mac-type="54455854"; x-mac-creator="646F7361"; name="BEKK.PRG"
Content-Transfer-Encoding: 7bit
Content-Description: PC-lukija Document
Content-Disposition: inline; filename="BEKK.PRG"

***  Estimating bivariate CAPM with GARCH(1,1); BEKK parametrization
*
calendar 1987 1 12
allocate 1997:4
open data hbhex.rat
data(format=rats) / hbor1 hbor2 hbor3 hbor6 hbor9 hbor12 hex
*
** data transformations: quarterly percentage rates, excess returns
*
set hex3 = (hex{-3}/hex)
set r3 = hbor3/400
set r6 = hbor6/400
set tb6s3 = ((1+r6)**2/(1+r3{-3}))-(1+r3)
set tbhex = (hex3)-(1+r3)
*
set U1 = 0.0
set U2 = 0.0
set b = 0.0
set H11 = 0.0
set H22 = 0.0
set H12 = 0.0
*
linreg(noprint) tb6s3 / R1
# constant
linreg(noprint) tbhex / R2
# constant
VCV(MATRIX=RR,noprint)
# R1 R2
*
nonlin C11 C12 C22 A11 A12 A21 A22 B11 B12 B21 B22 D1 D2
nlpar(criterion=coeffs)
FRML H11F = (VC=||C11,C12|0,C22||),$
(VA=||A11,A12|A21,A22||),$
(VB=||B11,B12|B21,B22||),$
(H=||H11{1}|H12{1},H22{1}||),$
(UB=||U1{1},U2{1}||*VB),$
(H=TR(VC)*VC+%MQFORM(H,VA)+TR(UB)*UB),$
H(1,1)
FRML H12F = H(1,2)
FRML H22F = H(2,2)
FRML regresid1 = tb6s3-D1*H12F
FRML regresid2 = tbhex-D2*h22F
FRML BETA = H12F/H22F
*
compute CINIT = %decomp(RR)
compute VC11 = CINIT(1,1),VC12 = CINIT(1,2),VC22 = CINIT(2,2)
*
FRML archlogl = (b(t)=beta(t)),(H11(t)=H11F(t)),(H12(t)=H12F(t)),(H22=H22F(t)),$
(U1(t)=regresid1(t)),(U2(t)=regresid2(t)),$
-.5*((log(%det(H)))+%qform(inv(H),UB))
*
stat(noprint) tb6s3
compute moy = %mean
compute C11 = %variance, A11 = 0.05, B11 = 0.0
stat(noprint) tbhex
compute D1 = moy/%variance
compute D2 = %mean/%variance
compute C22 = %variance, A22 = 0.05, B22 = 0.0
compute C12 = %cov(tb6s3,tbhex)
compute A12 = 0.00, A21 = 0.0, B12 = 0.0, B21 = 0.0
*
maximize(method=simplex,recursive,iteration=30) archlogl 2 *
maximize(method=bhhh,recursive,iteration=100) archlogl 2 *




--------------89026F74002D06D097E66A7A--


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

From: Jan Kakes <J.I.Kakes@eco.rug.nl>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: panel data
Date: Mon, 4 Aug 1997 11:53:49 GMT+0200
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
Organization:  Economische Faculteit RuG
X-mailer: Pegasus Mail v3.40 (via Mercury MTS (Bindery) v1.30)

RATS fellows,

I am using monthly panel data for a group of 9 countries, over a 
sample from 1970:01 up to 1995:12. My problem: how to tell RATS. 
If I use the following lines:

  calendar(panelobs=312) 1970 1 12
  allocate 9//1995:12
  open data panel.wk1
  data(format=wks,org=obs)

then RATS only reads the data of the first country, giving NA's to all
observations of the other countries. Constructing a file in RATS 
format, using RATSDATA, leads to the same problem. Everything works 
fine if I replace the first two lines by

  calendar(panelobs=312)
  allocate 9//312

and translate the results to a monthly frequency myself, but I still 
wonder if RATS can handle monthly data by itself in case of panel 
data, just like in the case of ordinary time series. Does anyone know 
how to do this?

Thanks,

Jan Kakes

University of Groningen
The Netherlands

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

From: "Estima" <estima@estima.com>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: panel data
Date: Mon, 4 Aug 1997 13:10:54 -0600
Errors-to: <rats-l-owner@efs.mq.edu.au>
Reply-to: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Sender: Maiser@efs1.efs.mq.edu.au
X-listname: <RATS-L@efs1.efs.mq.edu.au>
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
X-mailer: Pegasus Mail for Win32 (v2.42) (via Mercury MTS (Bindery) v1.30)

> I am using monthly panel data for a group of 9 countries, over a 
> sample from 1970:01 up to 1995:12. My problem: how to tell RATS. 
> If I use the following lines:
> 
>   calendar(panelobs=312) 1970 1 12
>   allocate 9//1995:12
>   open data panel.wk1
>   data(format=wks,org=obs)
> 
> then RATS only reads the data of the first country, giving NA's to all
> observations of the other countries. 

Jan:

RATS currently has no way of handling panel-type date information on
spreadsheet files. If you do have dates on your spreadsheet file
(i.e. a set of dates that repeat for each individual in the panel
set), RATS will only read the data for the first individual. I
suspect this is what is happening in your case.

If you don't have to worry about missing observations, I would 
suggest that you simply do the following:

calendar(panelobs=312)
allocate 9//312
open data panel.wk1
data(format=wks,org=obs)
calendar(panelobs=312) 1970 1 12

Specifying the "undated" panel CALENDAR should allow RATS to ignore 
the date information on the file, and thus read in all of the 
data. The second CALENDAR switches to a monthly/panel dating scheme, 
so that observations will be properly labeled in output, etc.

You could then store the data (along with the date information) to a 
RATS file by doing either:

open copy panel.rat
copy(for=rats,dates)

or:

dedit(new) panel.rat
store
save
quit

I hope this is helpful.

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

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

From: Peter Summers <p.summers@iaesr.unimelb.edu.au>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: panel data
Date: Tue, 05 Aug 1997 08:34:30 +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 Pro Version 2.2 (32) (via Mercury MTS (Bindery) v1.30)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"

Jan,
Another option (besides the one Tom suggested) is to create a RATS-format
data file and use the 'Form Panel' option in RATSDATA.  Then your example
code should work (adjusted for the right format, of course).



At 11:53 4/08/97 +0000, you wrote:
>RATS fellows,
>
>I am using monthly panel data for a group of 9 countries, over a 
>sample from 1970:01 up to 1995:12. My problem: how to tell RATS. 
>If I use the following lines:
>
>  calendar(panelobs=312) 1970 1 12
>  allocate 9//1995:12
>  open data panel.wk1
>  data(format=wks,org=obs)
>
>then RATS only reads the data of the first country, giving NA's to all
>observations of the other countries. Constructing a file in RATS 
>format, using RATSDATA, leads to the same problem. Everything works 
>fine if I replace the first two lines by
>
>  calendar(panelobs=312)
>  allocate 9//312
>
>and translate the results to a monthly frequency myself, but I still 
>wonder if RATS can handle monthly data by itself in case of panel 
>data, just like in the case of ordinary time series. Does anyone know 
>how to do this?
>
>Thanks,
>
>Jan Kakes
>
>University of Groningen
>The Netherlands
>
>
****************************************************************************
Peter Summers, Ph.D.
Research Fellow
Melbourne Institute 
  of Applied Economic and Social Research
University of Melbourne
Parkville, VIC 3052
AUSTRALIA
ph:  (03) 9344-5313
fax: (03) 9344-5630
email:  p.summers@iaesr.unimelb.edu.au
****************************************************************************


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

From: wan <wan@mail.tku.edu.tw>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: arfima and figarch!!
Date: Fri, 08 Aug 1997 17:23:30 +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 4.01 [en] (Win95; I) (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Are there anyone who knows the rats program or others can do
the estimation for arfima and fractional integrated model?thanks!!

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

From: David Montgomery <dmontgom@hawaii.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: arfima and figarch!!
Date: Fri, 8 Aug 1997 01:09:20 -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: Mozilla 4.01 [en] (Win95; I) (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Check out my home page listed in my sigmature file of the arfima program
that I wrote.  It duplicates the results of Sowell's paper for the
following arfima models.  (0,d,0)  (0,d,1) (0,d,2) (0,d,3) (1,d,0)
(1,d,1) (1,d,2) (1,d,3).  Let me know if you have any further questions.

wan wrote:

> Are there anyone who knows the rats program or others can do
> the estimation for arfima and fractional integrated model?thanks!!



--
____________________________________________________________________

David Montgomery
Ph.D. Candidate, Department of Economics
University of Hawaii at Manoa
Honolulu, Hawaii
E-mail: dmontgom@hawaii.edu
Home Page: http://www2.hawaii.edu/~dmontgom



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

From: David Montgomery <dmontgom@hawaii.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: arfima and figarch!!
Date: Fri, 8 Aug 1997 02:07:23 -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: Mozilla 4.01 [en] (Win95; I) (via Mercury MTS (Bindery) v1.30)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Check out the RATS homepage and the list of source code available.
Also, check out this homepage for the Ox program which has a complete
arfima program.
http://www.nuff.ox.ac.uk/Users/Doornik/doc/ox.htm#ox_over_pack


David Montgomery wrote:

> Check out my home page listed in my sigmature file of the arfima
> program
> that I wrote.  It duplicates the results of Sowell's paper for the
> following arfima models.  (0,d,0)  (0,d,1) (0,d,2) (0,d,3) (1,d,0)
> (1,d,1) (1,d,2) (1,d,3).  Let me know if you have any further
> questions.
>
> wan wrote:
>
> > Are there anyone who knows the rats program or others can do
> > the estimation for arfima and fractional integrated model?thanks!!
>
> --
> ____________________________________________________________________
>
> David Montgomery
> Ph.D. Candidate, Department of Economics
> University of Hawaii at Manoa
> Honolulu, Hawaii
> E-mail: dmontgom@hawaii.edu
> Home Page: http://www2.hawaii.edu/~dmontgom



--
____________________________________________________________________

David Montgomery
Ph.D. Candidate, Department of Economics
University of Hawaii at Manoa
Honolulu, Hawaii
E-mail: dmontgom@hawaii.edu
Home Page: http://www2.hawaii.edu/~dmontgom



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

From: "B. Dan Wood" <bdanwood@polisci.tamu.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: RE: arfima and figarch!!
Date: Fri, 8 Aug 1997 10:28:51 -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>
Organization: Political Science
X-Mailer: Microsoft Internet E-mail/MAPI - 8.0.0.4211 (via Mercury MTS (Bindery) v1.30)



-----Original Message-----
From:	wan [SMTP:wan@ntmail.tku.edu.tw]
Sent:	Friday, August 08, 1997 2:24 AM
To:	RATS Discussion List
Subject:	arfima and figarch!!

Are there anyone who knows the rats program or others can do
the estimation for arfima and fractional integrated model?thanks!!


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

From: "B. Dan Wood" <bdanwood@polisci.tamu.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: RE: arfima and figarch!!
Date: Fri, 8 Aug 1997 10:29:39 -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>
Organization: Political Science
X-Mailer: Microsoft Internet E-mail/MAPI - 8.0.0.4211 (via Mercury MTS (Bindery) v1.30)

There are programs for estimating ARFIMA up on RATS web page that can be downloaded.

-----Original Message-----
From:	wan [SMTP:wan@ntmail.tku.edu.tw]
Sent:	Friday, August 08, 1997 2:24 AM
To:	RATS Discussion List
Subject:	arfima and figarch!!

Are there anyone who knows the rats program or others can do
the estimation for arfima and fractional integrated model?thanks!!


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

From: "ABDOL S. SOOFI, ECONOMICS, UW-PLATTEVILLE; HTTP://VMS.WWW.UWPLATT.EDU/~SOOFI" <SOOFI@uwplatt.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: arfima and figarch!!
Date: Fri, 08 Aug 1997 12:09:36 -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 (Bindery) v1.30

Look up Geweke-Porter-Hudak routine in the RATS for estimating 
the ARFIMA model.

Good luck.




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

Abdol S. SOOFI                           
Department of Economics                  
University of Wisconsin-Platteville      
Platteville, WI 53818-3099 USA

(608) 342-1570
Soofi@uwplatt.edu
URL: http://vms.www.uwplatt.edu/~soofi

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

From: Christopher F Baum <baum@bc.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: arfima and figarch!!
Date: Fri, 08 Aug 1997 14:02:22 -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

The GPH routine (or Rob Schoen's Gaussian semiparametric routine of 
Robinson, also available from the Estima webpages) will estimate the 
fractional differencing parameter 'd', but they do not 'estimate an 
ARFIMA model' in the sense of, e.g., Sowell's exact maximum likelihood 
approach, in which p AR parameters and q MA parameters are jointly 
estimated with the d to allow for both short-term dependence (short 
memory) and long-term dependence (long memory). The earlier postings 
today regarding the sources of Sowell software mention some options in 
that regard. If the model is ARFIMA(p,d,q), GPH or Robinson estimators 
can be considered as estimating ARFIMA(0,d,0).

Kit Baum

On Fri, 08 Aug 1997 12:09:36 -0500 (CDT) "ABDOL S. SOOFI, ECONOMICS, 
UW-PLATTEVILLE; HTTP://VMS.WWW.UWPLATT.EDU/~SOOFI" <SOOFI@uwplatt.edu> 
wrote:

> Look up Geweke-Porter-Hudak routine in the RATS for estimating 
> the ARFIMA model.
> 
> Good luck.
> 
> 
> 
> 
> ***************************************************************************
> 
> ***************************************************************************
> 
> Abdol S. SOOFI                           
> Department of Economics                  
> University of Wisconsin-Platteville      
> Platteville, WI 53818-3099 USA
> 
> (608) 342-1570
> Soofi@uwplatt.edu
> URL: http://vms.www.uwplatt.edu/~soofi

---------------------------------------------------------------------
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: michaell@ORGO.CAD.GU.EDU.AU (Michael Lyons)
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: No Subject
Date: Mon, 11 Aug 1997 10:50:39 +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: Microsoft Mail via PostalUnion/SMTP for Windows NT (via Mercury MTS (Bindery) v1.30)
Mime-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: quoted-printable
Organization: Commerce Faculty, Griffith University, Australia


unsubscribe  =


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

From: Tomas Ekblad <tekblad@wasa.shh.fi>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: (no subject)
Date: Mon, 11 Aug 1997 19:50:33 -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: Mozilla 4.01 [en] (Win95; I; 16bit) (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: Steve L Beach <s_beach@mail.wsu.edu>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Kmenta Pooling
Date: Thu, 14 Aug 1997 17:28:07 -0700 (PDT)
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

Can someone tell me how the AR1 command is used (with DIFFERING?) to
calculate the Kmenta autoregressive, heteroskedastic, cross-sectionally
correlated model?  

I thought that was the approach, but the manual does not specify the
details, and it the AR1 does not give the cross-sectional correlations.

Otherwise, looks like it needs to be coded "from scratch."

All help appreciated.

Steven L. Beach
Department of Finance
College of Business and Economics
Washington State University
Pullman, WA 99164-4746

Phone: (509) 335-7822
E-mail: s_beach@mail.wsu.edu



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

From: d8341808@ms13.hinet.net
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Regime-Switching GARCH model
Date: Mon, 18 Aug 1997 10:03:16 +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: Windows Eudora Light Version 1.5.4 (32) (via Mercury MTS (Bindery) v1.30)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"

I want to  estimate a regime-switching GARCH model.
How can I use RATS commands to write this program. Could anyone help me?

Thanks in Advance

Institute of Business Adminstration
National Sun Tat-Swn University In Taiwan
E-mail D8341808@ms13.hinet.net


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

From: Jan <jan511@chinyi.ncit.edu.tw>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: An inquiry
Date: Thu, 22 Aug 1996 09:33:21 +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: 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

I am a user of RATS386 v.4.21. I wonder whether I can ask the following
question. The question happened when I run a trivariate GARCH whose
codes are
based on the Rob Trevor (1994). The codes are :
*
NONLIN DM0 DM1 DM2 DM3 DC0 DC1 A1 A2 AM BM CM
NONLIN(ADD) AC BC CC A1 B1 C1 AMC BMC CMC AM1 BM1 CM1 AC1 BC1 CC1
*
FRML REM = LNMR - DM0 - DM1*LNMR{1} - DM2*EX{1} - DM3*RINT{1}
FRML REC = CG - DC0 - DC1*CG{1}
FRML RE1 = EXP1 - (A1*VC1/VMC + A2*VM1/VMM) *  $
           (DM0+DM1*LNMR{1}+DM2*EX{1}+DM3*RINT{1}
*
FRML HMM = AM + BM*EM{1}**2 + CM*VMM{1}
FRML HCC = AC + BC*EC{1}**2 + CC*VCC{1}
FRML H11 = A1 + B1*E1{1}**2 + C1*V11{1}
FRML HMC = AMC + BMC*EM{1}*EC{1} + CMC*VMC{1}
FRML HM1 = AM1 + BM1*EM{1}*E1{1} + CM1*VM1{1}
FRML HC1 = AC1 + BC1*EC{1}*E1{1} + CC1*VC1{1}
*
DECLARE SYMMETRIC SIGMA
DECLARE VECTOR UVECT
FRML GLOGL = VMM(T) = HMM, VCC(T) = HCC, V11(T) = H11, $
             VMC(T) = HMC, VM1(T) = HM1, VC1(T) = HC1, $
             EM(T) = REM, EC(T) = REC, E1(T) = RE1, $
             SIGMA = ||VMM(T)|VMC(T),VCC(T)|VM1(T),VC1(T),V11(T)||, $
             UVECT = ||EM(T),EC(T),E1(T)||, $
             %LOGDENSITY(SIGMA,UVECT)
*
MAXIMIZE(METHOD=SIMPLEX,RECURSIVE,ITER=30) GLOGL GSTART GEND
MAXIMIZE(METHOD=BHHH,RECURSIVE,ITER=50) GLOGL GSTART GEND
*
The above is the main codes I used to estimate the estimates of GARCH.
However, I always run across the problem "Missing Values And/Or SMPL
Option Leave No Usable Data Points". Therefore, I printed the series
which
express in the FRML. The printed result showed that the following series
are reasonable: HMM,HCC,H11,HMC,HM1,HC1,REM,REC,RE1,EM,EC. But the
following series are
not available(NA): VMM,VCC,V11,VMC,VM1,VC1,E1,GLOGL. I don't know why
some are reasonable and the other are not. 

Yin-Ching Jan
Department of Industrial Engineering and Management
National Chin-Yi Institute of Technology
Taiwan, R.O.C.

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

From: gkoutmos@FAIR1.FAIRFIELD.EDU
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Re: An inquiry
Date: Fri, 22 Aug 1997 10:49:53 +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>
MIME-version: 1.0
X-Mailer: Pegasus Mail for Windows (v2.42a) (via Mercury MTS (Bindery) v1.30)
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT

It probably has to do with the way you define those series.  Set the 
variance vectors (not variance formulas) equal to the unconditional 
variance. It should work from that point on.


suppose v is the series that gives you NA and this is the vector that 
will hold the values estimated though the variance equation. Then you 


SET V = %VARIANCE

where %variance is the unconditional variance of the series under 
question

Good Luck



Greg Koutmos




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

From: Shikuan Chen <shikuan@haydn.ntu.edu.tw>
To: "RATS Discussion List" <RATS-L@efs.mq.edu.au>
Subject: Noncentral Chi-square Distribution
Date: Thu, 28 Aug 1997 15:19:53 +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 4.02 [en] (Win95; I) (via Mercury MTS (Bindery) v1.30)
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Hi!
    Is there any way to compute the noncentral chi-square distribution
from RATS?  Thanks in advance for any help!

Shikuan Chen
Department of International Business
National Taiwan University
Taipei Taiwan


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