Feature-weighted Stacking for Nonseasonal Time Series Forecasts: A Case Study of the COVID-19 Epidemic Curves

TL;DR

This paper explores a novel ensemble forecasting method combining statistical and deep learning models with meta-features, improving accuracy in nonseasonal COVID-19 epidemic curves for short-term predictions.

Contribution

It introduces a feature-weighted stacking approach using meta-features to enhance ensemble forecast accuracy for nonseasonal time series.

Findings

Meta-features improve forecast accuracy across horizons.

Combining Prophet and LSTM models yields better results.

Meta-features indicate the most predictive model for each case.

Abstract

We investigate ensembling techniques in forecasting and examine their potential for use in nonseasonal time-series similar to those in the early days of the COVID-19 pandemic. Developing improved forecast methods is essential as they provide data-driven decisions to organisations and decision-makers during critical phases. We propose using late data fusion, using a stacked ensemble of two forecasting models and two meta-features that prove their predictive power during a preliminary forecasting stage. The final ensembles include a Prophet and long short term memory (LSTM) neural network as base models. The base models are combined by a multilayer perceptron (MLP), taking into account meta-features that indicate the highest correlation with each base model's forecast accuracy. We further show that the inclusion of meta-features generally improves the ensemble's forecast accuracy across two forecast horizons of seven and fourteen days. This research reinforces previous work and demonstrates the value of combining traditional statistical models with deep learning models to produce more accurate forecast models for time-series from different domains and seasonality.