Online Ensemble Learning for Sector Rotation: A Gradient-Free Framework

TL;DR

This paper introduces a gradient-free online ensemble learning method for sector rotation that adaptively combines diverse models based on recent performance, improving prediction accuracy and robustness in financial forecasting.

Contribution

It develops a model-agnostic, gradient-free ensemble framework with theoretical performance guarantees, specifically tailored for nonstationary financial data and sector rotation applications.

Findings

Ensemble outperforms individual models and traditional methods in predictive accuracy.

Provides theoretical bounds on forecast regret based on out-of-sample R-squared.

Demonstrates robustness across different macroeconomic regimes, including COVID-19 crisis.

Abstract

We propose a gradient-free online ensemble learning algorithm that dynamically combines forecasts from a heterogeneous set of machine learning models based on their recent predictive performance, measured by out-of-sample R-squared. The ensemble is model-agnostic, requires no gradient access, and is designed for sequential forecasting under nonstationarity. It adaptively reweights 16 constituent models-three linear benchmarks (OLS, PCR, LASSO) and thirteen nonlinear learners including Random Forests, Gradient-Boosted Trees, and a hierarchy of neural networks (NN1-NN12). We apply the framework to sector rotation, using sector-level features aggregated from firm characteristics. Empirically, sector returns are more predictable and stable than individual asset returns, making them suitable for cross-sectional forecasting. The algorithm constructs sector-specific ensembles that assign adaptive weights in a rolling-window fashion, guided by forecast accuracy. Our key theoretical result bounds the online forecast regret directly in terms of realized out-of-sample R-squared, providing an interpretable guarantee that the ensemble performs nearly as well as the best model in hindsight. Empirically, the ensemble consistently outperforms individual models, equal-weighted averages, and traditional offline ensembles, delivering higher predictive accuracy, stronger risk-adjusted returns, and robustness across macroeconomic regimes, including during the COVID-19 crisis.