Direkt zum InhaltDirekt zur SucheDirekt zur Navigation
▼ Zielgruppen ▼

Humboldt-Universität zu Berlin -

Econometric Methods - Winter Term 2017/18

 

Lecturer: Prof. Dr. Bernd Droge

Course Material:
Further course information and material will ONLY be available on Moodle.

Lectures:
Mon, 10-12, SPA 1, 202
Tue, 12-14, SPA 1, 201

Exercises:
Thu, 14-16, SPA 1, 202
Fri, 12-14, SPA 1, 22

Content

1. Introduction

2. Principles of Linear Regression

  2.1. Basic Principles of Inference

       2.1.1. Econometric Modelling
       2.1.2. Basic Concepts
       2.1.3. Principles of Testing

  2.2. Ordinary Least Squares

       2.2.1. The Linear Regression Model
       2.2.2. The OLS Estimator
       2.2.3. Finite-Sample Properties
       2.2.4. Goodness-of-Fit
       2.2.5. Best Linear Prediction

  2.3. Hypothesis Testing under Normality

       2.3.1. Restricted Least Squares Estimation
       2.3.2. The F-Test
       2.3.3. The Chow Test

3. Generalizations and Applications of the Linear Model

  3.1. Selecting Regressors

       3.1.1. The Frisch-Waugh Theorem
       3.1.2. Functional Form and Marginal Effects
       3.1.3. Omitted and Irrelevant Variables
       3.1.4. Multicollinearity
       3.1.5. Dummy Variables

  3.2. Generalized Error Term Structures

       3.2.1. OLS with Generalized Errors
       3.2.2. The GLS Estimator
       3.2.3. Heteroscedasticity
       3.2.4. Autocorrelation

  3.3. Seemingly Unrelated Regression

       3.3.1. A System of Equations
       3.3.2. LS System Estimation

4. Asymptotic Properties of OLS

  4.1. Basic Concepts of Asymptotic Theory

       4.1.1. Modes of Convergence
       4.1.2. Convergence of Random Sequences
       4.1.3. Laws of Large Numbers and Central Limit Theorems

  4.2. Large-Sample Properties of OLS Estimation

       4.2.1. Asymptotic Distribution
       4.2.2. Special Cases
       4.2.3. Covariance Estimation
       4.2.4. Asymptotic Tests

5. Maximum Likelihood Estimation

  5.1. Basic Concepts

       5.1.1. The Likelihood Function
       5.1.2. The Maximum Likelihood Estimator

  5.2. Asymptotic Properties

       5.2.1. Regularity Conditions
       5.2.2. Consistency and Asymptotic Normality
       5.2.3. Pseudo ML

  5.3. ML Estimation of Linear Models

       5.3.1. The Homoscedastic Model
       5.3.2. The Heteroscedastic Model
       5.3.3. Other Examples

  5.4. Likelihood-Based Hypothesis Testing

       5.4.1. Restricted MLE
       5.4.2. Three Classical Tests
       5.4.3. Applications

  5.5. Numerical Optimization Methods

       5.5.1. Newton’s Method
       5.5.2. Gauss-Newton Regression

6. Instrumental Variables Estimation

  6.1. Introduction and Motivation

  6.2. IV Estimation of the Linear Model

       6.2.1. The Simple IV Estimator
       6.2.2. The Generalized IV Estimator
       6.2.3. Asymptotic Properties of the IV Estimator
       6.2.4. Two-Stage Least Squares
       6.2.5. Illustrations

  6.3. IV Based Testing

       6.3.1. Testing Overidentifying Restrictions
       6.3.2. The Hausman Test
       6.3.3. Choosing Instruments
       6.3.4. Weak Instruments

7. Generalized Method of Moments

  7.1. Method of Moments Estimation

       7.1.1. Conditional und Unconditional Moments
       7.1.2. The Method of Moments Estimator
       7.1.3. Special Cases of MM Estimation

  7.2. Generalized Method of Moments Estimation

       7.2.1. The GMM Estimator
       7.2.2. Asymptotic Properties of the GMM Estimator
       7.2.3. Optimal Choice of the Weighting Matrix
       7.2.4. Testing for Overidentifying Restrictions

Main Literature:
Davidson, R., and MacKinnon, J.G. (2004): Econometric Theory and Methods, Oxford University Press.
Hayshi, F (2000): Econometrics, Princeton University Press.


Additional Literature:
Heij, C., de Boer, P., Franses, P. H., Kloek, T., and van Dijk, H. K. (2004): Econometric Methods with Applications in Business and Economics, Oxford University Press.
Stock, J. H. and M. W. Watson (2003): Introduction to Econometrics, Boston, Mass. et al., Addison-Wesley.

 

Exam: written exam (150 min), homework assignments