The second module is an intermediate level course in applied econometrics. Topics covered include the specification and estimation of models pivoting on individual agents (firms, persons, etc.), as well as inference within these models. The main focus is estimation of models involving categorical and limited dependent variables, but students are also introduced to panel data techniques. The module begins with a basic exposition of panel data analysis covering 'fixed' and 'random' effects. It proceeds by reviewing the application of the linear probability model to the estimation of binary dependent variables. The alternative solutions offered by probit and logit models are then examined. Multiple choice models follow, specifically the ordered probit model and the multinomial logit model. The final part of the course addresses the estimation of limited dependent variables using tobit and sample selection models. For each of these topics, the models and concepts presented in the lecture are operationalized in lab classes using the STATA software.