Group Prediction in Information Markets With and Without Trading Information and Price Manipulation Incentives

The ability of individuals and groups to forecast a future event, with incomplete information, by using the trading history of an asset market is analyzed in the laboratory. The results show: (1) when forecasters observe the summary of markettransacted prices, they do not perform as well as when they are provided with a complete real-time sequence of bids, asks and contract prices; (2) groups do not outperform individuals in forecasting, and when the market does not have price manipulation incentives, individual prediction is better than the group prediction; (3) in markets with manipulators, where only a summary of contract prices is provided, both groups and individuals are unable to predict better than flipping a coin. This inability to aggregate information is remedied when forecasters see the complete evolution of market bids, asks and contracts.