Just Run Sweepstakes—Designing Prediction Markets to Counter Manipulation

• Main Authors: Meng-Chien Su, 蘇孟謙
• Other Authors: Joseph Tao-yi Wang
• Format: Others
• Language: en_US
• Published: 2017
• Online Access: http://ndltd.ncl.edu.tw/handle/86617967447688298081

碩士 === 國立臺灣大學 === 經濟學研究所 === 105 === We run the same prediction market with potential manipulation under four different incentive schemes: Real money, play money, contests and lotteries based on points earned. We find all markets work well when there is no manipulation, and bystanders adjust their forecasts according to evolving market price and excess bids/asks. However, when there is manipulation, forecasters rely more on evolving market price, but less on excess bids/asks (since manipulators can easily tamper them). Nonetheless, manipulators successfully mislead market prices under play money and contests because other traders have little incentives to react. In contrast, real money and lotteries provide incentive for arbitragers to counter manipulation, helping market prices (and forecasts) to aggregate information. This suggests using lotteries as legal alternatives to real money prediction markets.