Time series forecast of Covid 19 Pandemic Using Auto Recurrent Linear Regression

Abstract

Covid 19 pandemic has done severe impact in the economy and lifestyle of the people since the beginning of 2020. Various data analytics has been tried on the data obtained from various sources. These analytics include symptoms prediction, time series forecasting and impact analysis. The forecast on when the pandemic ends is a challenge for many countries. Time series forecasting models have been proposed for various applications but a non-seasonal and non-stationary forecasting method is needed to predict the progression of the pandemic. An Auto Regressive Linear Regression (ARLR) Algorithm is proposed in this paper with a selected geography’s Covid data. The results of the proposed methodology sounds convincing when compared to the non-seasonal and non-stationary existing methodologies like linear regression and exponential smoothing variants. The performance measure of standard deviation and RMSE of the proposed method obtained 430.22 and 0.31 for active cases while 27.01 and 0.77 for rate of transmission with positive skew and platykurtic trend.