Prediction Of New Covid Customers Using ARIMA Model

This paper aims to develop a model predicting weekly new COVID-19 cases in the near future for Bangladesh by analyzing past infection data. Three time series models were employed—general Exponential Smoothing, Holt’s Exponential Smoothing, and Autoregressive Integrated Moving Average (ARIMA)—with a focus on evaluating their accuracy and residual errors to determine the most effective model.