Application of harmony search algorithm in optimizing autoregressive integrated moving average: A study on a data set of Coronavirus Disease 2019

Abstract

World Health Organization (WHO) declared Coronavirus Disease 2019 (COVID-19), caused by the virus named Severe Acute Respiratory Syndrome Corona Virus-2 (SARS-CoV-2), as a global pandemic on March 11, 2020. Several researchers have used various statistical models and techniques to study and forecast the trend of spread and magnitude of the impact of SARS-CoV-2. ARIMA is one such model that has been widely used for this purpose, but in most cases, the model was not optimized suitably. In this paper, a music-inspired metaheuristic optimization algorithm, named Harmony Search (HS), has been integrated with ARIMA in order to improve the forecasting of the COVID-19 data set. The accuracy of forecasting has significantly increased after optimizing the model using the proposed algorithm. Furthermore, a novel application of HS has also been discussed in this paper.

Keywords

• Forecasting
• ARIMA
• Metaheuristics
• Harmony search
• Augmented Dickey–Fuller test
• COVID-19