Illusion of control as a source of poor diversification - Experimental evidence

This paper investigates factors that influence individual portfolio allocations with particular
focus on illusion of control. Participants in the experiment form their portfolio of two risky
lotteries and one risk-free alternative with the target to reach a predetermined income.
Subjects show illusion of control as they excessively invest in a lottery when they are in
charge of the chance move. This finding is amplified when self-selection is possible and
mitigated when a well-diversified default portfolio is offered. Presenting sequences of
chance moves prior to investment does not affect diversification. In line with excessive
extrapolation, the higher the number of observed positive prior outcomes, the more likely is
a positive prediction and in turn a higher investment.