Don't remove any variables if you want to preserve statistical properties of the analysis (coverage of CLs, relative unbiasedness of adjusted R-squared, P-values, unbiasedness of MSE).
Frank Harrell On Tue, 09 Apr 2002 00:42:30 GMT "Arthur Tabachneck" <[EMAIL PROTECTED]> wrote: > In either determining an R squared, or establishing a predictive equation, > should one remove dummy variable codings which indicate levels of a CLASS > variable from a model if the estimated parameter value is nonsignificant? > > Similarly, should one remove any variable, whether it be a level of a CLASS > variable or not, if the estimated parameter value is non-significant. > > Thanks in advance, > Art" > > > > -- Frank E Harrell Jr Prof. of Biostatistics & Statistics Div. of Biostatistics & Epidem. Dept. of Health Evaluation Sciences U. Virginia School of Medicine http://hesweb1.med.virginia.edu/biostat . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
