I'm trying to model market share (dependent variable) which is between 0 and 1. My boss told me to use Logistic Regression, but when I use the stepwise logistic regression with the backwards direction, my final model is only the intercept. Is this cause logistic regression should only be used when the dependent variable stricly 0 or 1? Is there anyway to model this scenario in R with the logistic regression while also using a step direction?
Thanks