English

Combining Predictive Distributions

Statistics Theory 2011-06-09 v1 Statistics Theory

Abstract

Predictive distributions need to be aggregated when probabilistic forecasts are merged, or when expert opinions expressed in terms of probability distributions are fused. We take a prediction space approach that applies to discrete, mixed discrete-continuous and continuous predictive distributions alike, and study combination formulas for cumulative distribution functions from the perspectives of coherence, probabilistic and conditional calibration, and dispersion. Both linear and non-linear aggregation methods are investigated, including generalized, spread-adjusted and beta-transformed linear pools. The effects and techniques are demonstrated theoretically, in simulation examples, and in case studies on density forecasts for S&P 500 returns and daily maximum temperature at Seattle-Tacoma Airport.

Keywords

Cite

@article{arxiv.1106.1638,
  title  = {Combining Predictive Distributions},
  author = {Tilmann Gneiting and Roopesh Ranjan},
  journal= {arXiv preprint arXiv:1106.1638},
  year   = {2011}
}
R2 v1 2026-06-21T18:19:36.282Z