Improving prediction market forecasts by detecting and correcting possible over-reaction to price movements

Abstract

We examine the impact of price trends on the accuracy of forecasts from prediction markets. In particular, we study an electronic betting exchange market and construct independent variables from market price (odds) time series from 6,058 individual markets (a dataset consisting of over 8.4 million price points). Using a conditional logit model, we find that a systematic relationship exists between trends in odds and the accuracy of odds-implied event probabilities; the relationship is consistent with participants overreacting to price movements. In particular, in different time segments of the market, increasing and decreasing odds lead, respectively, to under-and over-estimation of odds-implied probabilities. We develop a methodology to detect and correct the erroneous forecasts associated with these trends in odds in order to considerably improve the quality of forecasts generated in prediction markets.

Turning the heat on financial decisions: Examining the role temperature plays in the incidence of bias in a time-limited financial market

“…The greater the value of β, the greater the FLB. Consequently, β measures the degree to which betting odds under-/over-estimate the winning probabilities on favourites/longshots (Sung et al, 2019). To test whether β is statistically greater than 1, the standard normal test statistic is calculated:…”