Message-ID: <Pine.LNX.4.64.0602221351390.1081@gannet.stats.ox.ac.uk>
Date: 2006-02-22T13:53:53Z
From: Brian Ripley
Subject: does multinomial logistic model from multinom (nnet) has logLik?
In-Reply-To: <38b9f0350602220436v514bc933i@mail.gmail.com>
On Wed, 22 Feb 2006, ronggui wrote:
> I want to get the logLik to calculate McFadden.R2 ,ML.R2 and
> Cragg.Uhler.R2, but the value from multinom does not have logLik.So my
> quetion is : is logLik meaningful to multinomial logistic model from
> multinom?If it does, how can I get it?
>From the help page:
Value:
A 'nnet' object with additional components:
deviance: the residual deviance.
So it has a residual deviance. That is -2 log Lik in many cases (but not
if the argument 'summ' is used)
> Thank you!
>
> ps: I konw VGAM has function to get the multinomial logistic model
> with logLik, but I prefer use the function from "official" R
> packages .
>
> --
> ronggui
> Deparment of Sociology
> Fudan University
--
Brian D. Ripley, ripley at stats.ox.ac.uk
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UK Fax: +44 1865 272595