Optimal selection of forecasts

Many studies have shown that, in general, a combination of forecasts often outperforms the forecasts of a single model or expert. In this paper the authors postulate that obtaining forecasts is costly, and provide models for optimally selecting them. Based on normality assumptions, they derive a dynamic programming procedure for maximizing precision net of cost. The authors examine the solution for cases where the forecasters are independent, correlated and biased. They provide illustrative examples for each case.