keywords: COVID-19, ARIMA models, Holt’s linear exponential smoothing model, AIC
This paper evaluated and compared the performance of a family of smoothing models such as autoregressive integrated moving average (ARIMA) models andHolt’s exponential smoothing methods: Additive and multiplicative; to forecast the daily confirmed coronavirus (COVID-19) cases in Nigeria for the sampled period. The predictive capabilities were compared in terms of forecast accuracy measures, Akaike information criterion and Schwarz Bayesian information criterion based on the validated data set. The Holt’s linear exponential smoothing model with parameter and was found to have best described the data having the lowest ranked error statistics in an out of sample performance among other exponential smoothing models while the autoregressive integrated moving average (ARIMA (0, 1, 1)) with the smallest AIC was selected as the best. The forecast values of the two selected models show that the COVID-19 pandemic will live with us for a long term. The forecast results imply that the government and citizenry will and must adhere to preventive measures while going about their normal businesses.