Forecasting the spread of Covid-19 pandemic outbreak in India using ARIMA time series modelling.

COVID-19 is the infectious disease caused by the most recently discovered corona virus. This new virus and disease were unknown before the outbreak began in Wuhan, China, in December 2019. This paper focuses on a Time Series Model to predict COVID-19 Outbreaks in India. Every day data of fresh COVID-19 confirmed cases act as an exogenous factor in this frame. Our data envelops the time period from 01st Sep, 2020 to 9th Dec, 2020. COVID-19 Corona virus disease has been recognized as a worldwide hazard, and most of the studies are being conducted using diverse mathematical techniques to forecast the probable evolution of this outbreak. These mathematical models based on various factors and analyses are subject to potential bias. Here, we put forward a natural Times Series (TS) model that could be very useful to predict the spread of COVID-19. Here, a popular method Auto Regressive Integrated Moving Average (ARIMA) TS model is performed on the real COVID-19 data set to predict the outbreak trend of the prevalence and incidence of COVID-19 in.India. The time series under study is a non-stationary. Results obtained in the study revealed that the ARIMA model has a strong potential for prediction. The model predicted maximum COVID-19 cases in India at around 14, 22,337 with an interval (12, 80,352 - 15, 69, 817) during 1st Sep to 9th Dec period cumulatively. As per the model, the number of new cases shall fluctuate drastically in India. The results will help governments to make necessary arrangements as per the estimated cases. This kind of analysis and implications of ARIMA models and fitting procedures are useful in forecasting COVID-19 Outbreaks in India. [ABSTRACT FROM AUTHOR]