Prediction Markets at Yahoo! David Pennock Yiling Chen, Tej Kasturi, Havi Hoffman, Dan Reeves Chao-Hsien Chu, Sandip Debnath, Mike Dooley, Rael Dornfest, Joan Fiegenbaum, Gary Flake, Lance Fortnow, Brian Galebach, Lee Giles, Joe Kilian, Steve Lawrence, Bernard Mangold, Tracy Mullen, Rahul Sami, Emile Servan-Schreiber, Michael Wellman, Justin Wolfers
Prediction Y! *not to scale microeconomicsprediction markets
Prediction Y! DesignBuildAnalyze HSX NF TS WSEX FX PS
Case Study I Design Analyze Dynamic parimutuel –Market maker –Cross btw stock market and horse race betting Build
Parimutuel market
Dynamic parimutuel market C(1,2)=2.2 C(2,2)=2.8 C(2,3)=3.6 C(2,4)=4.5 C(2,5)=5.4 C(2,6)=6.3 C(2,7)=7.3 C(2,8)=8.2 C(3,8)=8.5 C(4,8)=8.9 C(5,8)=9.4 Share-ratio cost function: C(q 1,q 2 )= q 1 2 +q 2 2 Properties No arbitrage price i /price j = q i /q j price i < $1 payoff if right = C/q o > $1
Research Share-ratio price function One can view DPM as a market maker Cost Function: Price Function: Properties No arbitrage price i /price j = q i /q j price i < $1 payoff if right = C(Q final )/q o > $1
Yahoo!,OReilly launched Buzz Game Research testbed for investigating search futures Buy stock in hundreds of technologies Earn dividends based on actual search buzz API interface Exchange mechanism is new Yahoo!R invention Cross btw stock market and horse race betting
Technology forecasts iPod phone Google Calendar Another Apple unveiling 10/12; iPod Video search buzz price 9/8-9/18: searches for iPod phone soar; early buyers profit 8/29: Apple invites press to secret unveiling 8/28: buzz gamers begin bidding up iPod phone 9/7: Apple announces Rokr 9am 10/5
Forecast Accuracy Average forecast error across 352 stocks end of phase 1 contest period forecast error rapidly declines as traders zero in on correct predictions
Early Lessons Learned Forecasts point in right direction Market closing deadline focuses traders Dividend levels matter Intelligent strategies work –Randomized bots lost money to real traders –Contest winner followed optimal buzz trading strategy (prices buzz); Went from 4 th to 1 st place in final days –Forecast error does decrease over time
Analysis
Design | Delivery | Operations | Analysis Highly Customized Solutions Business-value markets since 2002, as covered in: Sample Customers « Trading uncertainty for collective wisdom » Prediction Market Competitive Forecasting TM Idea Pageant TM One Platform – Many Solutions
Whats next? What the buzz game thinks will grow –Web app framewks: Ruby on Rails –VOIP: Skype –Blinkx Desktop Search –Blog Services: WordPress –AJAXframeworks: scriptaculous –Linux: Ubuntu
Case Study II DesignBuild Yootopia: Y!Internal economy for favors, predictions, grp decisions Analyze
More Design Betting mechanisms –Combinatorics –Infinite/continuous –Linear programming backbone Market makers Applications BuildAnalyze Computational complexity Does money matter? Wisdom of crowds: Combining experts IOU social network Ledger system Buzz Game: Tech stock horse race