Weighted logistic regression

224 Views Asked by At

When we have a sample consisting of response variable has small number of 1 and large number of 0 (say 500 are 1 and 500000 are 0), is there a way to build a better logistic regression model using some methods? I googled and knew could put weights on the data but not so sure how. I am using R to build the model(glm()) so wonder if anyone could offer some guide.