THE REGRESSION TOURNAMENT: A NOVEL APPROACH TO PREDICTION MODEL ASSESSMENT

Main Article Content

Adi Schnytzer
Janez Janez Sustersic

Abstract

Standard methods to assess the statistical quality of econometric models implicitly assume there is only one person in the world, namely the forecaster with her model(s), and that there exists an objective and independent reality to which the model predictions may be compared. However, on many occasions, the reality with which we compare our predictions and in which we take our actions is co-determined and changed constantly by actions taken by other actors based on their own models. We propose a new method, called a regression tournament, to assess the utility of forecasting models and taking these interactions into account. We present an empirical case of betting on Australian Rules Football matches where the most accurate predictive model does not yield the highest betting return, or, in our terms, does not win a regression tournament.

Article Details

Section
Articles

References

McFadden, D. (1974). Conditional logit analysis of qualitative choice behavior. In P. Zarembka (Eds.), Frontiers in econometrics: Economic theory and mathematical economics (pp. 105-142). New York: Academic Press.

Ramanathan, R.,(2002), Introductory Econometrics with Applications 5th ed., South-Western.

Schnytzer, A. and G. Weinberg (2008), “Testing for Home Team and Favorite Biases in the Australian Rules Football Fixed Odds and Point Spread Betting Markets”, Journal of Sports Economics, Volume 9, No. 2, 173-190.

Schnytzer, A. (forthcoming), “The Prediction Market for the Australian Football League”, in L. Vaughn Williams, Prediction Markets, Routledge.

Zuber, R.A., Gandar, J.M., & Bowers, B.D. (1985), "Beating the spread: Testing the efficiency of the gambling market for national football league games". Journal of Political Economy, Volume 93, 800-806.