Logo Doctoral Certificate Program in Agricultural Economics Deutsch

Module 2500 in Hohenheim
Advanced Applied Econometrics:
Linear Models and Panel Data


Prof. Dr. Silke Hüttel, silke.huettel@uni-goettingen.de
Prof. Dr. Stefan Hirsch, s.hirsch@uni-hohenheim.de


At the end of the course, students shall

Skills: Methodological competence, quantitative analysis, conceptual thinking



Textbook: Verbeek, M. (2012): A Guide to Modern Econometrics, 4th edition. John Wiley & Sons
relevant chapters/sections or alternative readings given in parentheses

  1. Refresher Ordinary Least Squares (OLS) (2.1-2.6; 5.1)
    • The linear regression model
    • The Ordinary Least Squares Estimator (OLSE)
    • Properties of the OLSE (small sample and asymptotic properties)
    • Goodness of fit and hypothesis testing
  2. Refresher Generalized Least Squares (GLS) (4.1-4.3; 4.6)
    • Heteroskedasticity (introduction, implications for the estimator, testing, correction)
    • Autocorrelation (introduction, implications for the estimator, testing)
  3. Panel Data I: Static Linear Models for Panel Data (10.1-10.3)
    • Introduction to panel data
    • Models, assumptions and estimation
    • Testing
    • Goodness of fit
    • Two-way effects model
  4. Mixed Effects (Bates et al. 2015, Gardiner et al. 2009)
    • Random intercepts and random slopes
    • Nested effects versus crossed effects
    • Model selection and estimation issues
  5. Endogenous Regressors and Instrumental Variables Estimation (IVE) (5.1-5.4)
    • Review OLS properties and cases where OLS cannot be saved
    • The IV- and GIV-Estimator
    • Testing
    • Control Function Approach
  6. Weak Instruments (Stock and Yogo 2002)
    • Problem and detection of weak instruments
    • Testing for overidentifying restrictions
  7. Panel Data II: Instrumental Variables (10.4)
    • Panel data and endogenous regressors
    • Hausman-Taylor approach under correlated effects
  8. Generalized Method of Moments (GMME) (5.5-5.6)
    • The GMME
    • Illustration
    • Weak Identification
  9. Panel Data III: Dynamic Models for Panel Data (10.4-10.5; Bond 2002)
    • Model and assumptions
    • Estimation: Anderson-Hsiao, Arellano-Bond, Blundell-Bond System GMM
    • Testing
Teaching forms Workload (h)
1 week block seminar with lecture and exercise 40
Exercises 50
Total 90

Examination: Exercises

Grading: In-class and homework assignments, where a minimum score of 50% is required to pass the module.

Credit points: 3 CP

Language: English


Software: R

During the course all exercises can be conducted in R or Stata, while support will be given in R only and our handouts will be in R. Therefore, please familiarize yourself with basics in R if R is new to you. We recommend working at least through the following introductory material: