Prediction market

Prediction markets are virtual market platforms that predict the outcome of events. Prediction markets exist in the form of online betting exchanges or virtual securities markets, which are each implemented on an electronic platform and have their own quota or price determination mechanism. They are used as a competitive system to other forecasting tools.

• 4.2 Examples from industry

Development of the market forecast

For the first time the principle of forecast Marks arrived at the Iowa Electronic Markets in the U.S. state of Iowa, during the U.S. election in 1988, are used. The results of the IEM were more accurate than the forecasts of the market researchers.

Types of prediction markets

Virtual Securities Markets

No significant amounts of money or legal claims are traded on financial markets, in contrast to virtual securities markets. A phantom share represents a future event or a market condition (for example, sales of a product in the month of December or scored goals in a football game). The final value of the stock depends on each case from the actual outcome of the event, that is, for example, 1 ( virtual) € 100 per unit sales. Based on this relationship, then participants can trade their assessments. In contrast to stock market games, take the price of real exchanges, buy and sell orders on a prediction market its own trading mechanism are performed. Here rather the incentive system is essential to give the best prognosis. It is therefore also called gamification.

Betting Exchange

On -line betting exchanges placing participants betting on the outcome of sports competitions as well as social or political events. In contrast to virtual securities markets bet here with real sums of money on the outcome of events. A bet placed has the gain or the loss of real money result. A special contrast to traditional bet is for online betting exchanges is that the betting participants bet against each other rather than against a bookmaker.

Economics

A central explanation for the precision of prediction markets is the incentive system: players who buy low and sell high, so rewarded financially for improving prediction. Actors who buy expensive and sell cheap, so financially penalized for the deterioration of the prediction. This use of incentive mechanisms is also known as gamification.

The theoretical justification for the information efficiency of these markets provides the Hayek hypothesis, which states that in a market asymmetrically distributed information of market participants can be aggregated most efficiently by competition.

Companies use the principle of the market forecast or prediction markets to enable an exchange of relevant knowledge. How to strategic objectives will be achieved efficiently and effectively in order to build advantages over competitors. The exchange of knowledge in this form can provide management with a valuable source of knowledge. The more employees of the company belong to, the more untapped knowledge is present in it. It is this implicit knowledge of each employee can bring a prediction markets to give statements about uncertain scenarios or to dedicated issues. Here the future aspect is particularly important since so figures can be generated, which can be difficult to obtain from historical data.

Areas of application

Prediction markets can be used virtually anywhere where it comes to the prediction of uncertain events or where future-oriented answers, analyzes or forecasts to be delivered. Researchers emphasize the potential to improve policy and private sector decisions. To date, prediction markets from the U.S. Department of Defense for predicting terrorist attacks in the healthcare industry to predict flu outbreaks and the effectiveness of new drugs, as well as companies such as Eli Lilly, General Electric, Google, France Telecom, Hewlett Packard, IBM, Intel, Microsoft, Siemens, Zeppelin Rental, German Telekom and Yahoo used to predict sales figures or product qualities. Also, prediction markets could improve the accuracy of predictions of environmental disasters. Robin Hanson advocates Futarchie be laid down in the policy instruments on prediction markets.

That prediction markets can provide more accurate predictions than opinion polls, showed up at the 2005 federal election: Online election Political Stock Market stock market predicted that the CDU / CSU parliamentary group would win with 38.5 percent of the vote. This result is much closer to the official results of 35.2 percent than the average predicted 40 percent of the public opinion polls.

The results of the American Election Stock Market Iowa Electronic Market also show that prediction markets provide very reliable predictions. The four U.S. presidential elections from 1988 to 2000, the Iowa Electronic Market predicted share of the vote from Republicans and Democrats more accurate than the most reputable opinion polls. Predictions of the Iowa Electronic Market evaded 150 days before the election from only five percentage points from the official results. In the week immediately before the election, the prediction error was only 1.5 percent. In contrast, for example, the Gallup Institute erred in the most recent polls by an average of 2.1 percentage points.

Examples from industry

For example, prediction markets can be used to obtain sales forecasts and to estimate the potential of new product ideas. Companies implement this process on the intranet, thus saving costs and thus increase the security as well as when it is used externally. Thus offers the German market leader CrowdWorx successfully to this method.

Other examples are:

• Market forecasts, for example, the forecast of sales, market share, growth rates for any product. The consumer goods manufacturer Henkel uses Social Forecasting, thus raising its forecast accuracy by 22 percent.
• Competitive Intelligence, for example, risk of market entry of new competitors in the segment X estimate. Zeppelin Rental uses Social Forecasting to pool the knowledge of employees from its more than 100 locations for strategic issues.
• Product innovation, for example, to forecast flop rates, actual development costs and time to new product ideas. Tchibo has used Social Forecasting with the knowledge of his store employees for the evaluation of new products.
• R & D management, for example, the quantification of technology trends to assess development risks in good time. The German Telekom bundles the knowledge of some 240,000 employees in order to better assess the potential of new technologies.
• Economic forecasts, for example, unemployment rate, economic growth. The seed producer Syngenta uses social forecasting, early to adapt its production to the expected global demand.

Terms and legality

The use of prediction markets is limited by legal restrictions of each country. The core issue is the classification of the system as a game of skill as opposed to gambling. In each country apply to other legislation or the lack of a specific clarification on the issue of classification.

Renowned scientists such as Kenneth Arrow, Robert Forsythe, Robin Hanson, Saul Levmore, Paul Milgrom, Thomas Schelling, Robert Shiller, Vernon Smith, Cass Sunstein, Philip Tetlock, Hal Varian and Justin Wolfers call for a repeal of the statutory limitations because prediction markets have great potential to increase the social welfare.