R-users
E-mail: [email protected]
>My question is: Will this calculation be valid with the residual deviance
>returned by the glm() function using the quasibinomial family as
>reported in R?
Let me show you a simple example, assuming c=2.5:
function ()
{
xx <- c(1,2,3,4,5,6,7,8,9,10)
yy <- c(1,0,1,0,0,1,0,1,1,1)
data1 <- data.frame(x=xx, y=yy)
out1 <- glm(y~x, data=data1, family=binomial)
print(out1)
aic0 <- out1$aic
print("aic0")
print(aic0)
dev1 <- out1$deviance
aic1 <- dev1+ 2*2
print("aic1")
print(aic1)
c1 <- 2.5
qaic1 <- dev1/c1+ 2*2
print("qaic1")
print(qaic1)
}
The result is:
Call: glm(formula = y ~ x, family = binomial, data = data1)
Coefficients:
(Intercept) x
-0.7300 0.2131
Degrees of Freedom: 9 Total (i.e. Null); 8 Residual
Null Deviance: 13.46
Residual Deviance: 12.63 AIC: 16.63
[1] "aic0"
[1] 16.63054
[1] "aic1"
[1] 16.63054
[1] "qaic1"
[1] 9.052216
I hope that this R program will be of some help to you.
K. Takezawa
[[alternative HTML version deleted]]
______________________________________________
[email protected] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
