Predictive properties of forecast combination, ensemble methods, and Bayesian predictive synthesis

11/20/2019
by   Kosaku Takanashi, et al.
0

This paper studies the theoretical predictive properties of classes of forecast combination methods. The study is motivated by the recently developed Bayesian framework for synthesizing predictive densities: Bayesian predictive synthesis. A novel strategy based on continuous time stochastic processes is proposed and developed, where the combined predictive error processes are expressed as stochastic differential equations, evaluated using Ito's lemma. We show that a subclass of synthesis functions under Bayesian predictive synthesis, which we categorize as non-linear synthesis, entails an extra term that "corrects" the bias from misspecification and dependence in the predictive error process, effectively improving forecasts. Theoretical properties are examined and shown that this subclass improves the expected squared forecast error over any and all linear combination, averaging, and ensemble of forecasts, under mild conditions. We discuss the conditions for which this subclass outperforms others, and its implications for developing forecast combination methods. A finite sample simulation study is presented to illustrate our results.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
03/06/2018

Bayesian Predictive Synthesis: Forecast Calibration and Combination

The combination of forecast densities, whether they result from a set of...
research
04/05/2022

Aggregating distribution forecasts from deep ensembles

The importance of accurately quantifying forecast uncertainty has motiva...
research
03/10/2022

Spatially-Varying Bayesian Predictive Synthesis for Flexible and Interpretable Spatial Prediction

Spatial data are characterized by their spatial dependence, which is oft...
research
08/10/2023

Solving the Forecast Combination Puzzle

We demonstrate that the forecasting combination puzzle is a consequence ...
research
06/10/2021

Forecast combination based forecast reconciliation: insights and extensions

In a recent paper, while elucidating the links between forecast combinat...
research
01/28/2022

A loss discounting framework for model averaging and selection in time series models

We introduce a Loss Discounting Framework for forecast combination which...
research
06/08/2022

Bayesian Predictive Decision Synthesis

Decision-guided perspectives on model uncertainty expand traditional sta...

Please sign up or login with your details

Forgot password? Click here to reset