This contains help files for the class Econ 21130 at University of Chicago taught in the Winter of 2023. The class is taught by me, T. Lamadon. If you are a student, please sign up on the slack group.
The goal of the class is to get a better understanding of the mapping between models, data and the evaluation of realized and un-realized policies. The course is organized around environments with precisely defined data generating process derived from models using economic theory in the presence of randomness. Given a DGP will then think through the lens of a researcher who might not know all the components of the generating process but is given access to data and wants to create an informed prediction for the effect of a given policy intervention. We will follow this procedure for several environments which will each highlight different challenges and solutions.
The course focuses on micro-econometric methods that have applications to a wide range of economic questions. We study identification, estimation, and inference in both parametric and non-parametric models and consider aspects such as consistency, bias and variance of estimators. We discuss how repeated measurements can help with problems related to unobserved heterogeneity and measurement error, and how they can be applied to panel and network data. Topics include duration models, regressions with a large number of covariates, non-parametric regressions, and dynamic discrete choice models. Applications include labor questions such as labor supply, wage inequality decompositions and matching between workers and firms. Students will be expected to solve programming assignment in R.
For more details:
You will first need to install
Rstudio. To do use the following links:
You can install most packages directly: