Page 1 of 1
Standard Deviation of R^2
Posted: Sat Jul 16, 2011 6:24 am
by mwu888
Still not to0 familiar with the nomenclature in statistics. In Eviews, the output from a OLS regression shows standard error of regression. Can this be stated in plain English as the standard deviation of R^2? If not, how does one calculate the standard deviation of R^2?
Re: Standard Deviation of R^2
Posted: Sat Jul 16, 2011 7:47 am
by startz
The standard error of the regression is the estimated standard deviation of the error terms in the regression equation.
Out of curiousity, why do you want the standard deviation of the R-square?
Re: Standard Deviation of R^2
Posted: Sat Jul 16, 2011 6:59 pm
by mwu888
I read in an academic research paper in which the researchers compared R^2 in-sample versus R^2 out-of-sample and R^2 out-of-sample times standard deviation of R^2 to see the stability of the R^2. I am trying to understand this computation as well as the rationale behind it. Thank you.
Re: Standard Deviation of R^2
Posted: Sat Jul 16, 2011 7:24 pm
by startz
You might check "Mean and variance of R2 in small and moderate samples," by J.S. Cramer, Journal of Econometrics, July 1987.