[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: st: RE: VAR or VEC model

From   Ihtesham Afzal <>
To   <>
Subject   RE: st: RE: VAR or VEC model
Date   Tue, 25 Aug 2009 08:58:25 +0100

Thanks you for your reply.
But how then do I test for cointegration between the variables in the model? I was made to believe that the only way of finding cointegrating verctors was by doing the -vecrank- and -vec- commands. 
So do I use the var model and then -vargranger -to test for granger causality
and then -vec- and -vecrank- commands to test for cointegration?
Kind Regards

> Date: Tue, 25 Aug 2009 00:51:48 +0100
> Subject: Re: st: RE: VAR or VEC model
> From:
> To:
> Ihtesham Afzal wrote:
>> Thank you for your reply.
>> All of the variables are integrated of the same order (i.e I(1))
>> How do I run the granger causality test? Do I run the regression of one variable on the other with appropriate lags and then do an f-test to test whether these lags of the indepndent variable are equal to zero? And what command tdo I use to see what lag length to use -> I used -varsoc- is that right?
>> And then do this separately for all variables?
> -help vargranger-
> --
> Clive Nicholas
> [Please DO NOT mail me personally here, but at
> . Please respond to contributions I make in
> a list thread here. Thanks!]
> "My colleagues in the social sciences talk a great deal about
> methodology. I prefer to call it style." -- Freeman J. Dyson.
> *
> * For searches and help try:
> *
> *
> *
Windows Live Messenger: Thanks for 10 great years—enjoy free winks and emoticons.
*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index