Predicting the Impact of Covid-19 Pandemic in India

Darapaneni, N.; Maram, S.; Kour, M.; Singh, H.; Nagam, S.; Paduri, A. R.

COVID-19 is spreading at an unprecedented pace around the world. It has now spread to over 200 countries around the world. CoVID-19 mathematical modelling is often useful for strategic decision making in highly populated countries like India to gain some understanding of the epidemic's future. The prediction of an outbreak has become more difficult as the pandemic scenario of COVID-19 cases has grown. We hope to forecast the effects of COVID-19 in India by gaining a better understanding of its lifecycle in various Indian states. From historical data of verified COVID-19 cases, we are attempting to forecast potential COVID-19 cases and active cases.. For the prediction of COVID-19 we are implementing Susceptible-Infected-Recovered (SIR) model and FB-Prophet model for time series analysis. SIR modelling is more intuitive and explainable, but requires a lot of trial and error and assumptions. The FB-Prophet prediction process is simple and accuracy is also better compared to SIR modelling. In this model we are trying to understand the spread of COVID-19 in the ten most affected states of India (as on 9th December 2020) using publicly available state-wise time series data of COVID-19 patients. In this paper, we discuss how such continuous and unparalleled factors lead us to design intricate models, as It's time to use data-driven, mathematically proven models with the ability to tune parameters dynamically and automatically over time.