Suitable model prediction based on COVID 19 Phase I data

ABSTRACT

This paper integrates multiple standard regression models for prediction of COVID-19 infected data. We have taken Linear Regression, Polynomial Regression and Logistic Regression for our modelling and prediction purposes. These models are created, trialled and tested in MATLAB software with available data for Covid 19 infected cases. These models evolves as we get more and more data to show better predictions. Explanations of these models are valuable. The models’ forecasts are credible to epidemiologists and provide confidence in end-users such as policy makers and healthcare institutions as an output of this study. These models can be applied at different geographic resolutions, and in this paper, it is demonstrated for states in India. The model supplies more exact forecasts, in metrics averaged across the entire India. Lastly, we analyse the performance of our models for various datapoints and regression parameters to recommend optimized regression model.