Displaying 2 results from an estimated 2 matches for "m1big".
2009 Mar 17
2
bigglm() results different from glm()
...g code:
> require(biglm)
> options(digits=6, scipen=3, contrasts = c("contr.treatment",
"contr.poly"))
> dat=data.frame(y =c(rpois(50000, 10),rpois(50000, 15)),
ttment=gl(2,50000))
> m1 <- glm(y~ttment, data=dat, family=poisson(link="log"))
> m1big <- bigglm(y~ttment , data=dat, family=poisson(link="log"))
> summary(m1)
<snipped output for this email>
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) 2.30305 0.00141 1629 <2e-16 ***
ttment2 0.40429 0.00183 221 <2...
2009 Jul 03
2
bigglm() results different from glm()
Hi Sir,
Thanks for making package available to us. I am facing few problems if
you can give some hints:
Problem-1:
The model summary and residual deviance matched (in the mail below) but
I didn't understand why AIC is still different.
> AIC(m1)
[1] 532965
> AIC(m1big_longer)
[1] 101442.9
Problem-2:
chunksize argument is there in bigglm but not in biglm, consequently,
udate.biglm is there, but not update.bigglm
Is my observation correct? If yes, why is this difference?
Regards
Utkarsh
/
/
From: Thomas Lumley <tlumley_at_u.washington.edu
<mailto:tlu...