fETSmcs: Feature-based ETS model component selection

TL;DR

This paper introduces a fast, classifier-based method for selecting ETS model components, improving efficiency and accuracy in large-scale time series forecasting tasks.

Contribution

It proposes a novel classifier-based approach for ETS model component selection trained on simulated data, reducing computational complexity.

Findings

Effective model selection on simulated data

Improved forecasting accuracy on M4 dataset

Enhanced practical forecasting on hospital data

Abstract

The well-developed ETS (ExponenTial Smoothing or Error, Trend, Seasonality) method incorporating a family of exponential smoothing models in state space representation has been widely used for automatic forecasting. The existing ETS method uses information criteria for model selection by choosing an optimal model with the smallest information criterion among all models fitted to a given time series. The ETS method under such a model selection scheme suffers from computational complexity when applied to large-scale time series data. To tackle this issue, we propose an efficient approach for ETS model selection by training classifiers on simulated data to predict appropriate model component forms for a given time series. We provide a simulation study to show the model selection ability of the proposed approach on simulated data. We evaluate our approach on the widely used forecasting competition data set M4, in terms of both point forecasts and prediction intervals. To demonstrate the practical value of our method, we showcase the performance improvements from our approach on a monthly hospital data set.