Effects of overconfidence on decision making

This thesis aims to explore the effect of overconfidence on people's decision making. To approach this topic, a standard binary detection problem is considered, and its associated individual decision rule and decision fusion rule are derived. Following an axiomatic and empirical approach, a variant of the Prelec function from cumulative prospect theory is then developed to model the effect of overconfidence as a function of level of training. Next, the probability of detection after decision fusion is derived, and a combinatorial optimization is considered which aims to select a subgroup of people/agents to maximize the overall probability of detection.Ope