A Twitter-Based Prediction Market: Social Network Approach

This work develops an information system that combines the power of prediction markets with the popularity of Twitter and demonstrates that forecasting errors decrease with the cost of acquiring information in a network-embedded prediction market.

Abstract

Information aggregation mechanisms are designed explicitly for collecting and aggregating dispersed information. An excellent example of the use of this "wisdom of crowds" is a prediction market. The purpose of our Twitter-based prediction market is to suggest that carefully designed market mechanisms can elicit and gather dispersed information that can improve our predictions. We develop an information system that combines the power of prediction markets with the popularity of Twitter. Simulation results show that our network-embedded prediction market can produce better predictions as a result of the information exchange in social networks and can outperform other non-networked prediction markets. We also demonstrate that forecasting errors decrease with the cost of acquiring information in a network-embedded prediction market.