Abstract
Background: The second wave of coronavirus has appeared to be an extensive uphill of the number of daily new confirmed cases, recovered cases, and deaths than the first wave in India and the whole world. In India, the second wave of COVID-19 is much dangerous than the first wave that hit on 14th April 2020. The maximum number of new cases was 406901 recorded on May 7, 5.3 times more than the first wave peak. Many researchers worldwide are using machine learning prediction models to forecast the upcoming trends of this pandemic.
Methods: This study used an Autoregressive Integrated Moving Average (ARIMA) model to predict the daily new confirmed cases, daily new deaths, and daily new recoveries between and after the second wave of COVID-19 in India. The dataset was collected from March 14, 2020- July 7, 2021, using the ARIMA model to predict corona cases for the next 60 days.
Results: In the context of the current scenario in India, the second wave will score low new cases in mid-August 2021, and the third wave will hit the country in the middle of September 2021.
Conclusion: The ARIMA model was chosen based on AIC (Akaike Information Criterion) values and acquired the maximum accuracy of 95%.
Keywords: Covid-19, Second wave, third wave, Time series model, ARIMA, Forecast.
Graphical Abstract