GLM: Logistic Regression Fitted Probabilities Numerically 0 or 1 occurred for non-linearly separable data

Question

I have data that I don't believe is linearly separable. See below;

X = 761, 700, 3488, 555, 2784, 1336, 380

Y = 0, 1, 1, 0, 1, 1, 0

My belief is that because of the first two observations I shouldn't have linearly separable data. The error arises when I run:

glm(Y~X,family=binomial(link='logit'),data=data)

THanks in advance!

Answer 1

Good news. I get no such error and you're correct that the data are not linearly separable.

x=c(761,700,3488,555,2784,1336,380)
y=c(0,1,1,0,1,1,0)
f=glm(y~x,family=binomial)