subject:"\[R\] McFadden r\^2 and the inrercept"

Re: [R] McFadden r^2 and the inrercept

2011-10-25 Thread Joshua Wiley

Hi Ville, McFadden's pseudo R^2 is just a (log) likelihood ratio for your model compared to the null model. The null is typically an intercept only model, but other comparisons may be valid also. If memory serves: 1 - log(full_likelihood)/log(null_likelihood) you can often extract likelihoods

[R] McFadden r^2 and the inrercept

2011-10-25 Thread Ville Iiskola

Hi I have estimated parameters of my data with mlogit and the following commands. I would like to know also the McFadden R^2 and the intercept, could soweone tell me how that can be done? library(RODBC) library(mlogit) library(foreign) z<-odbcConnectExcel("D:\\MALLI11ARVOT.xls")