[R] Error in F test on version 2.1.0

[Prof Brian Ripley]

Why are you using an F test here?  The denominator is known, not 
estimated ....

On Mon, 9 May 2005, Ronaldo Reis-Jr. wrote:

> Em Seg 09 Mai 2005 10:53, Uwe Ligges escreveu:
>> Have you read the posting guide?
>> It asks you to specify a reproducible example ........
>> Please so so (both, read the psoting guide and specify the example).
>>
>> Uwe Ligges
>
>> tot <- c(16,20,23,18,10,18,25,20,24,24,29,12,23,23,21,23,11)
>> trat <- as.factor(rep(c("A","B"),c(8,9)))
>> dep <- c(0,1,1,0,1,0,1,1,15,13,12,4,12,10,9,10,5)
>> tot <- c(16,20,23,18,10,18,25,20,24,24,29,12,23,23,21,23,11)
>> m <- glm(dep/tot~trat,family=binomial,weights=tot)
>> anova(m,test="F")
> Analysis of Deviance Table
>
> Model: binomial, link: logit
>
> Response: dep/tot
>
> Terms added sequentially (first to last)
>
>
>     Df Deviance Resid. Df Resid. Dev      F Pr(>F)
> NULL                    16    105.685
> trat  1   96.117        15      9.568 96.117
> Mensagem de aviso:
> NaNs produzidos in: pf(q, df1, df2, lower.tail, log.p)

-- 
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