Overconfidence, ability and gender effects in anexperimental financial market

This thesis investigates individual’s overconfidence bias within the context of an experimental asset market. I relate overconfidence, gender differences and ability in judgement to individual’s trading performance and trading activities. The most robust finding that was found in the psychology of judgment issue, with people found to be generally overconfident and evidence of this overconfidence seen in several disciplines. In the financial market, the theoretical models predict that overconfident investors trade excessively. Consequently, trading too much is hazardous to investors’ wealth and performances. To contribute to the emerging literature on this topic, overall, 376 participants were involved in an experimental study, through which their degree of overconfidence in judgement have been measured using subjective confidence interval estimation (miscalibration test). In addition to misclaribration score, other psychological bias measures and control variables are also included. During the financial experiments, those individuals were allowed to trade with the virtual agents or against other participants given certain endowments of both cash and shares, depending on the trading rules of each experimental market. By varying the market structure, trading data was collected which that enables us to obtain a much better picture in relation to the actual individuals’ behaviour in this experimental environment.