st: Getting Separate Chi-Squared Index for Each Lags in VAR Model

[James Zhang <latent_hydralisk@hotmail.com>]

Hi everyone,

I would appreciate it if someone could help me with this problem.I am trying to use VAR model to test for correlations between four markets: Shanghai, Hongkong, Taiwan and S&P 500, with 1-4 lags. And I am stuck with the Chi2 index. This is according to one of the Statalist guys in Boston:

tsset _0
qui var shashr hsi twse_index s_p_500,lag(1/4)
forv i=1/4 {
testparm (L`i'.shashr L`i'.hsi L`i'.twse_index L`i'.s_p_500)}

But it doesn't work. Each regression should have four chi-squared statistics, i.e., one for the lags on SH, one for the lags on HK, one for the lags on TW, and one for the lags on the S&P500.  That is, it is like an F-test for the lags on each variable in each regression.  Thus, each VAR with multiple lags selected will yield a total of 16 chi-squared statistics (given four dependent variables times four sets of lags in each regression).
Please kindly share your wisdom on how to do this in STATA. Thanks.



Sincerely Yours,
James
 		 	   		  
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/