Estimating Collective Belief in Fixed Odds Betting
Fixed odds betting is a popular mechanism in sports game betting. In this paper, we aim to decipher actual group belief on contingent future events from the dynamics of fixed odds betting. Different from previous studies, we adopt the prospect theory rather than the expected utility (EU) theory to model bettor behaviors. Thus, we do not need to make assumptions on how much each bettor stake on their preferred events. We develop a model that captures the heterogeneity of bettors with behavior parameters drawn from beta distributions. We evaluate our proposed model on a real-world dataset collected from online betting games for 2008 Olympic Game events. In the empirical study, our model significantly outperforms expert (bookmaker) predictions. Our study shows the possibility of developing a light-weight derivative prediction market upon fixed odds betting for collective information analysis and decision making.
Keywordsfixed odds betting prediction markets computational experiments
Unable to display preview. Download preview PDF.
- 2.Schrieber, J.M.: The Application of Prediction Markets to Business. Engineering Systems Division (2004)Google Scholar
- 7.Erik, S., Justin, W.: Explaining the Favorite-Long Shot Bias: Is it Risk-Love or Misperceptions? In: Chicago, IL, ETATS-UNIS, vol. 118, pp. 723–746. University of Chicago Press, Chicago (2010)Google Scholar
- 10.Justin, W., Eric, Z.: Interpreting Prediction Market Prices as Probabilities. Institute for the Study of Labor, IZA (2006)Google Scholar
- 15.Chen, Y., et al.: Information markets vs. opinion pools: an empirical comparison. In: 6th ACM Conference on Electronic Commerce. ACM Press, Vancsouver (2005)Google Scholar