Long Run Predictions Using Gompertz Curves - A State Wise Analysis of COVID-19 Infections in India

The aim of this paper is to perform a State wise Analysis of the First and the Second COVID- 19 Waves experienced by India using the Gompertz Curves and to estimate the maximum number of affected individuals for each wave with the best possible accuracy. A total of 21 large States are chosen for the analysis encompassing 97% of the Indian population. Data on cumulative number of cases is available till 31st October 2021. The entire dataset is segregated into two parts, i.e., the First and the Second Waves and then modelled individually by the Gompertz Curves with some generalizations. The predicted maximum cumulative numbers of COVID-19 affected individuals are found to be quite accurate. Besides, it is found to be possible to give a methodology how one can predict these numbers with a much smaller dataset. This is important as it can help the authorities in taking an informed decision on the efficient allocation of the limited health care resources.

Keywords:COVID-19, disease modeling, Gompertz Curve, Non-linear least squares, time series, Forecasting, Prediction
JEL Classifications: E0, C32, C50, C53