Chapter 4.1 Logistic Regression

Question 1

Logistic Regression
Models of discrete choice have been a topic in (Micro-)Econometrics and are nowadays widely used in Marketing.

Logit, and probit models extend the principles of general linear models (regression) to better treat the case of …

They focus on categorical dependent variables, looking at all …

… is also known as classification.

McFadden got the 2000 Nobel prize in Economics for fundamental contributions in discrete choice modeling.

Answer 1

dichotomous and categorical target variables.

levels of possible interaction effects.

Predicting a categorical dependent variable

Question 2

What does it mean that error terms are heteroscedastic?

Answer 2

The predicted probabilities of the linear model can be greater than 1 or less than 0. 𝜀 is not normally distributed because 𝑌 takes on only two values.