Misestimating Probability Distributions of Repeated Events

Oleg UrminskyUniversity of Chicago Booth School

Abstract

This paper examines people’s subjective beliefs about probability distributions arising from repeated events, such as the number of heads in ten coin flips. Across elicitation methods and decision scenarios, people express beliefs that are systematically biased relative to the actual distribution, over-estimating the tails and under-estimating the shoulders of the distribution. While experts are relatively more accurate than novices, both show significant bias.

Files

Misestimating Probability Distributions of Repeated Events (398 KB)



Back to Table of Contents