Coronavirus disease (COVID-19) is an infectious disease caused by a coronavirus that is circulating worldwide. Various countries have used various measures to combat the disease's spread. Many studies have adopted the mathematical modeling to predict the cases during the pandemic. In our study we have used Box- Jenkins’s Auto Regressive Integrated Moving Average (ARIMA) time series mathematical model. MATERIALS AND METHODS: Publicly available data of daily COVID-19 confirmed cases along with Meteorological variables were considered using Expert Modeler in SPSS to predict and forecast COVID19 cases in Delhi region, India. RESULTS: Spearman’s correlation was used to find the relationship between COVID-19 cases with Meteorological variables. Humidity, rainy days and Average sunshine were found to be significant. ARIMA (0, 1, 14) model found to be best fitted model for the given data with R square value of fitted model is 0.920. Ljung-Box test value is 39.368 with p value showing significant, indicating that the fitted model is adequate to predict and forecast COVID19 cases. CONCLUSION: ARIMA (0, 1, 14) mathematical model was selected as a best suited model to predict and forecast the incidence of COVID-19 cases in Delhi region, which would be useful for the policymakers for better preparedness.

Keywords : COVID-19, Mathematical Model, ARIMA Model, Meteorological Variables, Ljung-Box test