[R] p-value in lrm

[Frank E Harrell Jr]

donghu at itsa.ucsf.edu wrote:
> Hi,
> 
> How can I get the p-value, S.E., and Z from the result of lrm fit?  Thank you.
> 
> Donglei

That's automatic.  The print method for lrm in the Design package does 
that.  Better though is the anova method which gives you the right 
pooled tests when you have nonlinear effects, interactions, etc.

If by "get" you mean "retrieve for use in other code" you can look at 
print.lrm for hints, or use coef(fit) and diag(Varcov(fit)).

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