On Mon, Dec 19, 2011 at 10:24:18AM -0500, [email protected] wrote: > On Mon, Dec 19, 2011 at 10:19 AM, Jieyun Fu <[email protected]> wrote: > > It's Wald's Z test in here. Search for Wald Test in
> > http://userwww.sfsu.edu/~efc/classes/biol710/logistic/logisticreg.htm > > In ordinary linear regression they like to call it t-statistics. I guess > > it's just a terminology thing. > Since the focus of scikits.statsmodels is more on parameter > estimation, maybe you should get the statistics from us. Yes, I agree. The logistic regression from the scikit-learn is meant to be used in 'penalized' settings. This means that it is a biased estimator, and thus it is much harder to get p-values on the parameters estimated, as we are doing only a point estimate, and not computing the full posterior. For hypothesis testing, it does seems that using statsmodels would be more natural. Gael ------------------------------------------------------------------------------ Write once. Port to many. Get the SDK and tools to simplify cross-platform app development. Create new or port existing apps to sell to consumers worldwide. Explore the Intel AppUpSM program developer opportunity. appdeveloper.intel.com/join http://p.sf.net/sfu/intel-appdev _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
