This study constitutes the first comprehensive examination of Canadian mutual fund performance using a dataset free of all conditioning biases. The goal is to test many of the same hypotheses which have been previously addressed using US data. The sample is carefully constructed so as to avoid not only survivorship bias but also a form of backfilling bias that exists because funds have a timing option as to when to first provide results to information vendors. The deleterious impact of both forms of bias is documented. Not unlike what has been found in the US, on average fund managers net-of-expenses underperform benchmarks, but it also seems clear that their analysis and trading contribute to portfolio performance. I also present evidence that, at least on a short-term basis, success breeds success. Investors seem aware of this since money flows to successful funds. The strategy of chasing returns looks to be a viable one. One useful byproduct of this work is that an independent dataset has allowed for the corroboration of many of the same stylized facts that have been previously observed in the US.

We perform an asset market experiment in order to investigate whether overconfidence induces trading. We investigate three manifestations of overconfidence: calibration-based overconfidence, the better-than-average effect and illusion of control. Novelly, the measure employed for calibration-based overconfidence is task-specific in that it is designed to influence behavior. We find that calibration-based overconfidence does engender additional trade, though the better-than-average also appears to play a role. This is true both at the level of the individual and also at the level of the market. There is little evidence that gender influences trading activity.

In 12 sessions conducted in a typical bubble-generating experimental environment, we design a pair of assets that can detect both irrationality and speculative behavior. The specific form of irrationality we investigate is the probability judgment error associated with low-probability, high-payoff outcomes. Independently, we test for speculation by comparing prices of identically paying assets in multiperiod versus single-period markets. We establish that aggregate irrationality measured in one dimension (probability judgment error) is associated with aggregate irrationality measured in another (bubble formation).

In 12 sessions conducted in a typical bubble-generating experimental environment, we design a pair of assets that can detect both irrationality and speculative behavior. The specific form of irrationality we investigate is the probability judgment error associated with low-probability, high-payoff outcomes. Independently, we test for speculation by comparing prices of identically paying assets in multiperiod versus single-period markets. We establish that aggregate irrationality measured in one dimension (probability judgment error) is associated with aggregate irrationality measured in another (bubble formation).