A Statistics and Deep Learning Hybrid Method for Multivariate Time Series Forecasting and Mortality Modeling

TL;DR

This paper introduces MES-LSTM, a hybrid multivariate time series forecasting method combining statistical and deep learning techniques, demonstrating superior accuracy and uncertainty quantification on COVID-19 datasets.

Contribution

The paper presents MES-LSTM, a novel multivariate extension of ES-RNN that addresses computational and dependency challenges in hybrid forecasting models.

Findings

MES-LSTM outperforms pure statistical and deep learning methods in accuracy.

The hybrid approach provides more reliable prediction intervals.

Consistent improvements observed across multiple COVID-19 datasets.

Abstract

Hybrid methods have been shown to outperform pure statistical and pure deep learning methods at forecasting tasks and quantifying the associated uncertainty with those forecasts (prediction intervals). One example is Exponential Smoothing Recurrent Neural Network (ES-RNN), a hybrid between a statistical forecasting model and a recurrent neural network variant. ES-RNN achieves a 9.4\% improvement in absolute error in the Makridakis-4 Forecasting Competition. This improvement and similar outperformance from other hybrid models have primarily been demonstrated only on univariate datasets. Difficulties with applying hybrid forecast methods to multivariate data include ($i$) the high computational cost involved in hyperparameter tuning for models that are not parsimonious, ($ii$) challenges associated with auto-correlation inherent in the data, as well as ($iii$) complex dependency (cross-correlation) between the covariates that may be hard to capture. This paper presents Multivariate Exponential Smoothing Long Short Term Memory (MES-LSTM), a generalized multivariate extension to ES-RNN, that overcomes these challenges. MES-LSTM utilizes a vectorized implementation. We test MES-LSTM on several aggregated coronavirus disease of 2019 (COVID-19) morbidity datasets and find our hybrid approach shows consistent, significant improvement over pure statistical and deep learning methods at forecast accuracy and prediction interval construction.