Volume 7 Issue 4 ( December 2020 )


Predictive Models and Analysis of Peak and Flatten Curve Values of CoVID-19 Cases in India

Paras Bhatnagar, Shivendra Kaura, Sanjeev Rajan


Worldwide increasing cases of COVID-19 are putting high pressure on healthcare services. The coronavirus epidemic caused announcing emergency cases in India. The virus started with one infected case by 30th January, 2020, in Kerala, where the first death due to corona in Karnataka and 73 announced cases were reported by 12th March, 2020. 1024 announced cases were reported by 29th March, 2020.This indicates that the number of confirmed cases is increasing rapidly, causing national crises for India. This study aims is to fill a gap between previous studies and the current development of COVID-19 spreading, by extracting a relationship between corona positives as independent and corresponding deaths as a dependent variable. This research statistically analyses the mortality in 10 days of every month. A mathematical model to predict the new deceased cases corresponding infected cases in a practical scenario is proposed. An approximate prediction of mortality corresponding to new predicted cases can be easily performed using the proposed model. As most of the other countries have reached their peaks, confirmed cases start decreasing. By analyzing these countries’ data considering the lockdown, the peak ratio is identified using all countries’ population and the decreasing rate of confirmed cases after the peak has been achieved. The same calculation has been done for death and recovered cases. This average peak ratio is used to identify India’s COVID patients’ peak value. The decreasing rate is also used to define the rate of confirmed cases and mortality after the peak has occurred. The model has also been verified in different countries to identify the significance of the model.

Keywords: COVID-19, Correlation, regression, a test of significance, Machine Learning.