Econometrics: Estimation Methods, SS 2021, Mark Trede

Econometrics: Estimation Methods

This course provides an introduction to the main estimation methods used in econometrics and their application in Python. The course aims are (i) to understand the theoretical basis of the most important estimation methods, (ii) to be able to apply the methods to new models, (iii) to learn the numerical methods required to perform the estimations.

Contents

• Brief introduction to Python
• Maximum likelihood estimation: basic idea; score vector; information matrix; general properties; covariance matrix estimation; dependent observations; the classical tests
• Generalized method of moments (GMM): classical method of moments; basic idea of GMM; general properties; covariance matrix estimation; test of over-identifying restrictions; method of simulated moments
• Indirect inference: basic idea; auxiliary models
• Quantile regression
• Bootstrapping: basic idea; parametric and nonparametric approach; smoothed bootstrap; model based bootstrap; confidence intervals; hypothesis tests

Administration

• The course belongs to the methods module (“Forschungsmethoden”) of the PhD program. It may also earn a certificate (A, B, or C) for other doctoral students.
• Maximum number of participants: 15.  
• Master student may attend if there are vacancies.
• The course (including the exam) is eligible for 6 credit points.
• The exam consists of a take-home exercise (80%) and a 30-minute written exam (20%).
• Prerequisites: Basic knowledge in statistics and time series analysis. Prior knowledge of basic computer programming (e.g. loops) is helpful, prior knowledge of “Python” is not necessary.