[R] reporting multinomial logistic regression results

Bob Green bgreen at dyson.brisnet.org.au
Thu Apr 12 13:15:26 CEST 2007

I am hoping to obtain information regarding the reporting of 
multinomial logistic regression results - either recommendations of 
published studies employing R or suggestions about what output should 
be reported. I am used to SPSS which routinely produces a broader 
range of output than R (e.g wald statistic p values, exp(B) and 
associated confidence intervals).

Below is the code I have been using for a series of analyses.

mod.multacute <- multinom(offence ~ in.acute.danger * 
violent.convictions, data = kc,  na.action = na.omit)
summary(mod.multacute, cor=F, Wald=T)
Anova (mod.multacute)
predictors <- expand.grid(in.acute.danger = c("y","n"), 
violent.convictions = c("y","n"))
p.fita <- predict(mod.multacute, predictors, type='probs')

Any assistance is appreciated,


Bob Green

More information about the R-help mailing list