Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | mantoshkinas <mantoshkinas@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | st: Re: How to perform a Vuong test in STATA v10 on Panel Data in Fixed effects model? |
Date | Wed, 11 May 2011 13:41:58 +0200 |
How to perform a Vuong test in STATA v10 on Panel Data in Fixed effects model? I need to compare 2 similar types of regressions and the literature in my field (Accounting & Finance) indicates that it's good to use the Vuong test (Vuong Z-statistics or likelihood ratio test for model selection) to compare which of those 2 regressions has a statistically higher R-squared (not by just comparing the absolute difference between R-squareds of the models). STATA has the Vuong test but it is used only in binomial, negative binomial and Poisson regressions. My models and the models in those studies that use Vuong do not contain binomial variables as the dependent variables. In particular, my dependent variable in both regressions is annual Stock (Security) Returns of a company and the independent variables in one regression are Accounting Earnings (measured as Net Income) & change in Accounting Earnings (change in Net Income) and Cash Flow from operations and the change in the cash flow from operations in the other one. I'd be so grateful if someone could indicate how to compare R-squareds in those regressions using a statistical test rather than just looking at the absolute difference between R-squared with bare eye. Sincerely- Mantoshkinas * * 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/