Saved in:
Bibliographic Details
Main Authors: Medeiros, Marcelo C., Pinro, Jeronymo M.
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2508.20795
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914011095760896
author Medeiros, Marcelo C.
Pinro, Jeronymo M.
author_facet Medeiros, Marcelo C.
Pinro, Jeronymo M.
contents The forecasting combination puzzle is a well-known phenomenon in forecasting literature, stressing the challenge of outperforming the simple average when aggregating forecasts from diverse methods. This study proposes a Reinforcement Learning - based framework as a dynamic model selection approach to address this puzzle. Our framework is evaluated through extensive forecasting exercises using simulated and real data. Specifically, we analyze the M4 Competition dataset and the Survey of Professional Forecasters (SPF). This research introduces an adaptable methodology for selecting and combining forecasts under uncertainty, offering a promising advancement in resolving the forecasting combination puzzle.
format Preprint
id arxiv_https___arxiv_org_abs_2508_20795
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Time Series Embedding and Combination of Forecasts: A Reinforcement Learning Approach
Medeiros, Marcelo C.
Pinro, Jeronymo M.
Econometrics
The forecasting combination puzzle is a well-known phenomenon in forecasting literature, stressing the challenge of outperforming the simple average when aggregating forecasts from diverse methods. This study proposes a Reinforcement Learning - based framework as a dynamic model selection approach to address this puzzle. Our framework is evaluated through extensive forecasting exercises using simulated and real data. Specifically, we analyze the M4 Competition dataset and the Survey of Professional Forecasters (SPF). This research introduces an adaptable methodology for selecting and combining forecasts under uncertainty, offering a promising advancement in resolving the forecasting combination puzzle.
title Time Series Embedding and Combination of Forecasts: A Reinforcement Learning Approach
topic Econometrics
url https://arxiv.org/abs/2508.20795