A Statistics and Deep Learning Hybrid Method for Multivariate Time Series Forecasting and Mortality Modeling
1
School of Computer Science and Applied Mathematics, University of the Witwatersrand, Johannesburg 2000, South Africa
2
Institute for Intelligent Systems, University of Johannesburg, Johannesburg 2092, South Africa
*
Author to whom correspondence should be addressed.
†
These authors contributed equally to the conceptualisation and presentation of this work.
Academic Editors: Sonia Leva, Francesca Perla and Salvatore Scognamiglio
Forecasting 2022, 4(1), 1-25; https://doi.org/10.3390/forecast4010001
Received: 24 October 2021
/
Revised: 12 December 2021
/
Accepted: 15 December 2021
/
Published: 22 December 2021
(This article belongs to the Special Issue Mortality Modeling and Forecasting)
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.
View Full-Text
Keywords:
deep learning; multivariate time series forecasting; prediction intervals; mortality modeling
▼
Show Figures
Figure 1
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited
MDPI and ACS Style
Mathonsi, T.; van Zyl, T.L. A Statistics and Deep Learning Hybrid Method for Multivariate Time Series Forecasting and Mortality Modeling. Forecasting 2022, 4, 1-25. https://doi.org/10.3390/forecast4010001
AMA Style
Mathonsi T, van Zyl TL. A Statistics and Deep Learning Hybrid Method for Multivariate Time Series Forecasting and Mortality Modeling. Forecasting. 2022; 4(1):1-25. https://doi.org/10.3390/forecast4010001
Chicago/Turabian StyleMathonsi, Thabang, and Terence L. van Zyl. 2022. "A Statistics and Deep Learning Hybrid Method for Multivariate Time Series Forecasting and Mortality Modeling" Forecasting 4, no. 1: 1-25. https://doi.org/10.3390/forecast4010001
Find Other Styles
