Prediction markets for electricity demand

Load forecasting is a central task for operating, maintaining, and planning power systems. Because of this importance, many different methods are proposed to forecast load, but none of them is proved clearly superior. This paper proposes a prediction market to forecast electricity demand that has the advantage of allowing aggregation and competition among the many available methods. We describe how to implement a simple prediction market for continuous variables, using only contracts based on binary variables. We also discuss possible pitfalls in the implementation of such a market.