Reference no: EM131131885

You want to estimate whether there is a relationship between gender and smoking at your university. You collect data for a random sample of 200 students and run a linear probability model, a probit model and a logit model, where the dependent variable is a binary indicator for whether an individual smokes, and the main explanatory variable is a dummy variable indicating males (you also include other controls such as age). Your estimated coefficients are 0.084 in the LMP, 0.688 in the probit model, and 0.389 in the logit model. In spite of these large differences, all three models yield similar estimates of the marginal effect of gender on the probability of smoking. How can this be?