Ensemble Modeling for Time Series Forecasting: an Adaptive Robust Optimization Approach

TL;DR

This paper introduces an adaptive robust optimization method for ensemble time series forecasting, which dynamically adjusts model weights to improve accuracy and robustness across various real-world applications.

Contribution

It presents a novel adaptive robust optimization approach for ensemble modeling that outperforms existing techniques in time series forecasting tasks.

Findings

Adaptive ensembles outperform individual models by 16-26% in RMSE.

The method improves conditional value at risk by 14-28%.

Demonstrated effectiveness across synthetic and real-world datasets.

Abstract

Accurate time series forecasting is critical for a wide range of problems with temporal data. Ensemble modeling is a well-established technique for leveraging multiple predictive models to increase accuracy and robustness, as the performance of a single predictor can be highly variable due to shifts in the underlying data distribution. This paper proposes a new methodology for building robust ensembles of time series forecasting models. Our approach utilizes Adaptive Robust Optimization (ARO) to construct a linear regression ensemble in which the models' weights can adapt over time. We demonstrate the effectiveness of our method through a series of synthetic experiments and real-world applications, including air pollution management, energy consumption forecasting, and tropical cyclone intensity forecasting. Our results show that our adaptive ensembles outperform the best ensemble member in hindsight by 16-26% in root mean square error and 14-28% in conditional value at risk and improve over competitive ensemble techniques.