[R] Help understanding lrm function of Design library

Frank E Harrell Jr f.harrell at vanderbilt.edu
Tue Aug 18 14:13:02 CEST 2009

Noah Silverman wrote:
> Hi,
> I'm developing an experiment with logistic regression.
> I've come across the lrm function in the Design  library.
> While I understand and can use the basic functionality, there are a ton 
> of options that go beyond my knowledge.
> I've carefully read the help page for lrm, but don't understand many of 
> the arguments and optional return values.
> (penalty, penalty.matrix, strata.penalty, var.penalty, weights)
> Additionally, there are several items returned in the model that I don't 
> understand.
> (Max Derriv, Model L.R, C, Dxy, Gamma, Tau-a, R2, Briar, Wald Z)
> 1) Could someone point me to an online resource where I could learn 
> more?  (I'm big on trying to teach myself.)
> 2) Or, alternately, does anybody want to explain a few things for me?
> Thanks!
> -- 
> Noah

My book Regression Modeling Strategies covers much of that.  Its case 
study on penalized ordinal logistic regression is also largely in a 
paper in Statistics in Medicine if you have access to that.  Once you do 
some background reading I'll be glad to answer some focused questions.

Frank E Harrell Jr   Professor and Chair           School of Medicine
                      Department of Biostatistics   Vanderbilt University

More information about the R-help mailing list