puzzling results from logistic regression

Michael <comtech.usa <at> gmail.com> writes:

Hi all,

As you can see from below, the result is strange...

I would imagined that the bb result should be much higher and close to 1,
any way to improve the fit?

Any other classification methods?

Thank you!

data=data.frame(y=rep(c(0, 1), times=100), x=1:200)
aa=glm(y~x, data=data, family=binomial(link="logit"))

newdata=data.frame(x=6, y=100)
bb=predict(aa, newdata=newdata, type="response")
bb

bb

1

0.4929125

I have a feeling you meant to say

data <- data.frame(y=rep(c(0,1), each=100), x=1:200)

instead. 

Try 

with(data,plot(y~x))

for each data set to see what you actually got as opposed to
what you thought you were getting it.

You may still have a little bit of a problem fitting such an
extreme data set -- this is what is called "complete separation",
and leads to an infinite estimate of the slope -- if you want
to pursue this, take a look at the brglm package.