Abstract:
We compare the performance of financial professionals (CFAs) with university students in four financial forecasting tasks ranging from simple lab prediction tasks to longitudinal field tasks. Although students and professionals performed similarly in the artificial forecasting tasks, their performance differed in the more realistic tasks. Yet, increasing the 'representativeness of the situation' in the lab tasks did not systematically benefit financial professionals as students outperformed CFAs when forecasting historical series. However, professionals outperformed students in the field task. Our results imply that the expertise of financial professionals might have been underestimated in previous works that focused on lab tasks.