Command for GRS test in STATA

[priya sawaliya]

Respected sir,
I have time series data on 25 portfolios excess return (dependent variable) and 6 independent variables (RM-RF, SMB, HML, RMW, CMA AND WML). I want to run GRS test to check whether these 25 intercepts are jointly zero. I have run these 25 regressions with command "foreach y of varlist y1-y25 {regress `y' RM-RF, SMB, HML, RMW, CMA AND WML}". I have tried a lot but not able to do this GRS test. Please help me...
Thank you
Priya

[Phil Bromiley]

You didn't get a quick answer. You'll increase your chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex (which you do). Note also that most of us are not from your area - you'll increase your chances of a useful answer by explaining in general terms.

I have no idea what a GRS test is. You can test whether intercepts from multiple regressions are zero by either (i) using sureg, or (ii) stack the data so you can run all the regressions in one regression where you use dummies to allow different parameters for each of the 25 portfolios and the explanatory variables.

[Cliff Zhu]

There's a new package "grsftest" doing GRS test. It does not adjust for the degrees of freedom when calculating estimator of the sample covariance matrix of the factor portfolios. This approach avoids a common misrepresentation of the GRS paper.

ssc install grsftest

[Joro Kolev]

Another way is to implement the GRS test through -[MV] manova-.

Yes another way is to implement the GRS test through -mvreg- and to adjust the degrees of freedom manually.

In both cases the GRS test amounts to simply testing that the intercepts in the multivariate regression are jointly zero.