An ensemble approach to forecast COVID-19 incidences using linear and non-linear statistical models. (15th January 2022)

• Record Type: Journal Article
• Authors: Mahajan, Asmita; Sharma, Nonita; Almukhtar, Firas Husham; Mangla, Monika; Sharma, Krishna Pal; Rani, Rajneesh
• Is Part Of: International journal of computer applications technology. Volume 66:Number 3/4(2021)
• Page Start: 415
• Page End: 426
• Publication Date: 2022-01-15
• Subjects: coronavirus -- COVID-19 -- pandemic -- forecasting -- ensemble -- stacking -- auto regressive integrated moving average -- exponential smoothing -- neural network -- multi-layer perceptron; Technology -- Data processing -- Periodicals; 620.00285
• Languages: English
• ISSNs: 0952-8091

Abstract

Coronavirus has infected many countries across the globe and is still spreading. As health sectors are experiencing an unexpected rise in cases each day, researchers are perpetually trying to discover a permanent antidote for the virus. Considering the spread, it becomes imperative to predict the spread of disease to mitigate the spread. This paper proposes a stacked ensemble model to predict occurrences of COVID-19 using Exponential Smoothing (ETS), Auto Regressive Integrated Moving Average (ARIMA), and Neural Network Auto-regression (NNAR) as the base models. Each base model is trained individually on the disease data set, whose regress values are then used to train the Multi-Layer Perceptron (MLP) model. The stacked ensemble model outperforms the base models in terms of Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). The results conclude that the ensemble model is competent to forecast future COVID incidences in comparison to other statistical time series models.