Statistical Analysis of Novel COVID-19 Based on Real-Time Data and Future Epidemics

Sekhar, C. H.; Srinivasa Rao, M.; Battacharyya, D.

The buzzword in the health sector globally was COVID-19 alias coronavirus. Present all over the world, the virus spread rapidly from one person to another. The outbreak was first reported in Wuhan, China, in December 2019, exponentially spreading to the entire China. In the past months, COVID-19 becomes an outbreak throughout the world. The impact of this new virus brought horror to several countries as cities are quarantined and locked down and hospitals are overcrowded. A high infectious ailment that compromises the worldwide world, part of things isn’t completely comprehended. Our study focuses on COVID-19-infected cases in India. Sickness classification is one of the significant and more time-consuming tasks in medical diagnosis system. In India, COVID-19 cases are slowly increasing day by day. Compared to advanced countries, the spread is under control. Here, we have customized our prepared dataset based on the government of India’s data provided by ICMR. Statistical analysis on COVID-19-infected, death and recovered cases among India’s various states is based on the real-time data. To forecast the coronavirus’s epidemic peak, what might help us act appropriately to reduce the epidemic risk? We used compartmental model in epidemiology, the SEIR method, to forecast the number of confirmed cases based on the current scenario and population. Further, we used machine learning methods such as ARIMA and SEIR models to forecast India’s confirmed cases in the coming days and to start to forecast the decrease of infected cases by the end of the year 2020. Our study can predict the infected patients based on symptoms using ARIMA and SEIR model. At the last comparison of the cases based on prediction and with lockdown cases. Finally, we suggested preventive measures to minimize COVID-19 infections. This work expects to predict and forecast COVID-19 cases, deaths and recoveries through predictive modelling. The model helps to interpret public sentiment patterns on scattering related health information and assess the political and financial impact of the spread of the infection. Methods: Real-time data query has been visualized in this work. The queried data is used for susceptible-exposed-infectious-recovered (SEIR) predictive modelling and forecasting the number of cases and infected using autoregressive integrated moving average (ARIMA). We utilized SEIR model and ARIMA model to forecast COVID-19 outbreak within India based on daily observations. Findings: At the time of writing this book chapter, the number of confirmed cases is expected to exceed 473,000 and reach the peak of this outbreak before 30 June 2020. This outbreak is assumed to peak in late August 2020 and will start to drop around early September 2020. © 2022, Springer Nature Switzerland AG.