Forecast Aggregation

27 Pages Posted: 18 Mar 2017

See all articles by Itai Arieli

Itai Arieli

Technion-Israel Institute of Technology

Yakov Babichenko

Technion, Industrial Engineering and Managemenet

Rann Smorodinsky

Technion-Israel Institute of Technology - The William Davidson Faculty of Industrial Engineering & Management

Date Written: March 16, 2017

Abstract

Bayesian experts with a common prior that are exposed to different types of evidence possibly make contradicting probabilistic forecasts. A policy maker who receives the forecasts must aggregate them in the best way possible. This is a challenge whenever the policy maker is not familiar with the prior nor the model and evidence available to the experts. We propose a model of non-Bayesian forecast aggregation and adapt the notion of regret as a means for evaluating the policy maker's performance. Whenever experts are Blackwell ordered taking a weighted average of the two forecasts, the weight of which is proportional to its precision (the reciprocal of the variance), is optimal. The resulting regret is equal 1/8(5*sqrt(5)-11)≈0.0225425, which is 3 to 4 times better than naive approaches such as choosing one expert at random or taking the non-weighted average.

Keywords: Forecast Aggregation, Aggregation Scheme

Suggested Citation

Arieli, Itai and Babichenko, Yakov and Smorodinsky, Rann, Forecast Aggregation (March 16, 2017). Available at SSRN: https://ssrn.com/abstract=2934104 or http://dx.doi.org/10.2139/ssrn.2934104

Itai Arieli (Contact Author)

Technion-Israel Institute of Technology ( email )

Technion City
Haifa 32000, Haifa 32000
Israel

Yakov Babichenko

Technion, Industrial Engineering and Managemenet ( email )

Hiafa, 3434113
Israel

Rann Smorodinsky

Technion-Israel Institute of Technology - The William Davidson Faculty of Industrial Engineering & Management ( email )

Haifa 32000
Israel

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
160
Abstract Views
993
Rank
334,454
PlumX Metrics