| ![]() |
From | "Gauri Khanna" <gwkhanna@hotmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: White test for heteroskedasticity: maxvar |
Date | Thu, 17 Aug 2006 16:22:06 +0000 |
From: Rudy Fichtenbaum <rudy.fichtenbaum@wright.edu>_________________________________________________________________
Reply-To: statalist@hsphsun2.harvard.edu
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: White test for heteroskedasticity: maxvar
Date: Thu, 17 Aug 2006 08:05:16 -0400
Gauri,
You might also consider running the special case of White's test suggested by Wooldridge. He suggests regressing uhat squared (the squared residual) on yhat and yhat squared (the predicted value from an ols regression and the predictied value squared). The predicted value and the predicted value squared are functions of all of your independent variables. This amounts to imposing some restrictions on the function that contains all of the original independent variables, the independent variables squared and the cross products. You can calculate the LM statistic by taking n times the R square from the equation above which has a chi square distribution with two degrees of freedom.
Rudy
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
© Copyright 1996–2025 StataCorp LLC | Terms of use | Privacy | Contact us | What's new | Site index |