External validity of prospect theory: The evidence from soccer betting

Abstract

This paper tests whether the prospect theory parameters estimated from laboratory experiments correspond to estimates from real-life betting markets. Using Czech soccer betting data, we estimate the functional forms of the value and probability weighting functions commonly used for the experimental validation of prospect theory. In line with the experimental evidence, we find that bettors are risk averse in the domain of gains and risk seeking in the domain of losses and tend to overweight small probabilities and underweight medium and large probabilities. On the other hand, our findings suggest that bettors do not exhibit loss aversion. This might be at least partly explained by recent experimental evidence suggesting that loss aversion is weaker or absent if the decision-makers are experienced, if they face similar choices repeatedly, and if the decisions are made in a context where offers are usually accepted.