Multinomial logistic regression with nnet library

["Inman, Brant A. M.D." <Inman.Brant@mayo.edu>]

I am using S-Plus 7.0 to model a polychotomous response variable.  Using the multinom() function of the nnet library, I am able to get the following fit:

> fit.mn
Call:
multinom(formula = me ~ hist)

Coefficients:
  (Intercept)  hist
1      -0.951 1.257
2      -1.250 1.009

Residual Deviance: 792.30
AIC: 800.3

However, I don't know how to get the standard errors for the regression coefficients generated by this function.  I have tried the summary function and get the following message:

> summary(fit.mn)
Problem in svd.default(X): x must be numeric or complex
Use traceback() to see the call stack

Any ideas?

_________________________________________________
Dr. Brant A. Inman,  MD FRCSC
Urologic Oncology Fellow, Mayo Clinic College of Medicine
200 First Street SW
Rochester, MN   55905
Phone: (507) 284-1330
Fax:    (507) 284-4951
Email:  inman.brant@mayo.edu
_________________________________________________