Re-interpreting Prediction Market Price as Aggregated Probabilities

摘要

Prediction markets (PMs) have gained successes in predicting the future events probabilities. Several studies have argued why markets can aggregate the collective belief and reflect this collective belief in PMs' price. These existing works based their arguments on the assumptions about how markets participant react to market price, e.g. expected utility theory (EU). In this work wefirst translate the equilibrium price into fixed odds betting term. Thus we study how the mean belief of the bettors influences the final share of bettor on each side in binary fixed odds betting based on more reasonable consumerrnbehavior models, e.g. Prospect Theory (PT) and some other rational assumptionon consumer's heterogeneity.We established computational experiments for ourmodel and summarize the experiment results in a compact mathematical formula. From the formula we can conclude that the market price is a good approximate to the mean belief of all the market participants.