Master-Seminar Summer 2018

Econometric Pitfalls in Regression Analysis: by means of Growth


A wide variety of papers are available presenting empirical evidence for determinants of economic growth. Depending on the specific focus variables measuring political stability, human resources, technology, capital accumulation, credit ratings and many other factors are identified as “key drivers” of economic growth. Contrary to the range of results those papers have the empirical strategy in common – regression analysis on cross-section or panel dataset.

In this seminar we will investigate common pitfalls for this type of analysis. We start with a real research paper[1] and then check the robustness of the results with respect to potential econometric pitfalls.

Here exemplary a list of papers concerning those pitfalls:

(1) Extreme-Bounds Analysis

Levine, Ross, and David Renelt. "A Sensitivity Analysis of Cross-Country Growth Regressions." The American Economic Review 82, no. 4 (1992): 942-63.

Xavier X. Sala-I-Martin. "I Just Ran Two Million Regressions." The American Economic Review 87, no. 2 (1997): 178-83.

(2) Spurious Regression

Chatelain, Jean-Bernard, and Kirsten Ralf. „Spurious regressions and near-multicollinearity, with an application to aid, policies and growth“ Journal of Macroeconomics 39, Part A (2014): 85-96.

Kao, Chihwa. „Spurious regression and residual-based tests for cointegration in panel data.“ Journal of Econometrics 90, Issue 1 (1999): 1-44

(3) Outlier and Leverage

Fox, John. „Regression diagnostics: An introduction (Vol. 79)“. Sage (1991)

Chatterjee, Samprit, and Ali S. Hadi. "Influential Observations, High Leverage Points, and Outliers in Linear Regression." Statistical Science 1, no. 3 (1986): 379-93.


Time Schedule:

Thur. May   3rd  2018;   2.00 – 4.00 pm.; HS V          Kick-off event

Thur. May 24th  2018;   2.00 – 6.00 pm.; HS V          Round table (discussion and exchange)

Mon. June 26th 2018;  12.00 am.                              Submission final term paper

Thur. June 28th 2018;    2.00 – 6.00 pm.; Software-Lab (01-115; ReWi II)   Presentation



This seminar requires a sound knowledge of statistical concepts, especially regression analysis using cross-section and panel data and some basic knowledge in time series analysis. The discussed methods have to be applied to real data so practical experiece with a statistic toolbox like Stata, Matlab or R is expected.

Further information will be provided in the Kick-off event.


[1]Barro, Robert J. and Jong-Wha Lee. “Source of economic growth”. Carnegie-Rochester Conference Series on Public Policy 40 (1994): 1-46.