# Time Series Embedding and Combination of Forecasts: A Reinforcement Learning Approach

## TL;DR

This paper introduces a reinforcement learning framework for dynamic forecast model selection and combination, aiming to overcome the forecasting combination puzzle by improving aggregation performance on diverse datasets.

## Contribution

It presents a novel reinforcement learning-based method for adaptive forecast combination, addressing the longstanding challenge of outperforming simple averaging methods.

## Key findings

- Effective in simulated and real data scenarios
- Improves forecast accuracy over traditional methods
- Demonstrates robustness across datasets like M4 and SPF

## Abstract

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.