Limitations of Ensemble Bayesian Model Averaging for Forecasting Social Science Problems

12 Pages Posted: 18 May 2013 Last revised: 30 Aug 2014

See all articles by Andreas Graefe

Andreas Graefe

Macromedia University of Applied Sciences

Helmut Kuechenhoff

Ludwig Maximilian University of Munich (LMU)

Veronika Stierle

Technische Universität München (TUM) - Department of Economics

Bernhard Riedl

Technische Universität München (TUM) - Department of Economics

Date Written: August 30, 2014

Abstract

A reanalysis and extension of Montgomery, Hollenbach, and Ward (2012) shows that the predictive performance of Ensemble Bayesian Model Averaging (EBMA) strongly depends on the conditions of the forecasting problem. EBMA is of limited value in situations with small samples and many component forecasts, a situation that is common for social science prediction problems. These results conform to a large body of research, which has determined that simple approaches to combining (such as equal weights) often perform as well as sophisticated approaches when combining forecasts. Simple averages are easy to describe, easy to understand, and easy to use. They should be favored over more complex methods unless one has strong evidence that differential weights will improve accuracy.

Keywords: combining forecasts, equal weights, election forecasting

JEL Classification: C1

Suggested Citation

Graefe, Andreas and Kuechenhoff, Helmut and Stierle, Veronika and Riedl, Bernhard, Limitations of Ensemble Bayesian Model Averaging for Forecasting Social Science Problems (August 30, 2014). International Journal of Forecasting, Forthcoming. Available at SSRN: https://ssrn.com/abstract=2266307 or http://dx.doi.org/10.2139/ssrn.2266307

Andreas Graefe (Contact Author)

Macromedia University of Applied Sciences ( email )

Sandstrasse 9
Munich, Bavaria 80337
Germany

HOME PAGE: http://www.andreas-graefe.org

Helmut Kuechenhoff

Ludwig Maximilian University of Munich (LMU) ( email )

Geschwister-Scholl-Platz 1
Munich, Bavaria 80539
Germany

Veronika Stierle

Technische Universität München (TUM) - Department of Economics ( email )

Ludwigstr. 28
Munich, 80539
Germany

Bernhard Riedl

Technische Universität München (TUM) - Department of Economics ( email )

Ludwigstr. 28
Munich, 80539
Germany

Register to save articles to
your library

Register

Paper statistics

Downloads
105
Abstract Views
1,082
rank
255,167
PlumX Metrics