ARIMA Modelling and Forecasting of COVID-19 Daily Confirmed/Death Cases: A Case Study of Nigeria
Essi Isaac Didi
Department of Mathematics, Rivers State University, Port Harcourt, Rivers State, Nigeria.
Nwuju Kingdom
*
Department of Mathematics, Rivers State University, Port Harcourt, Rivers State, Nigeria.
Etuk Ette Harrison
Department of Mathematics, Rivers State University, Port Harcourt, Rivers State, Nigeria.
*Author to whom correspondence should be addressed.
Abstract
Aims: The aim of this work is to develop suitable ARIMA models which can be sued to forecast daily confirmed/death cases of COVID-19 in Nigeria. This is subject to developing the model, checking them for suitability and carrying out eight months forecast, and making recommendations for the Nigerian Health sector.
Study Design: The study used daily confirmed and death cases of COVID-19 in Nigeria.
Methodology: This work covers times series data on the on the daily confirmed/death cases of COVID-19 in Nigeria, obtained from the Nigerian Centre for Disease Control (NDCD) from 21 March 2020 to 5 May 2020, covering a total of 51 data points. This work is geared towards developing a suitable Autoregressive Integrated Moving Average (ARIMA) models which can be used to forecast total daily confirmed/death cases of COVID-19 in Nigeria. Two adequate subset ARIMA (2, 2, 1) and AR (1) models for the confirmed/death cases, respectively, is fitted and discussed
Results: A forecast of 239 days – from 6th May 2020 to 31 December 2020 was conducted using the fitted models and we observed that the COVID19 data has an upward trend and is best forecasted within a short period.
Conclusion: Critical investigation into the rate of spread of COVID-19 pandemic has shown that, that the daily confirmed cases as well as death cases of the disease tends to follow an upward trend. This work aimed at developing a suitable ARIMA models which can be used to fit a most appropriate subsets to statistically forecast the actual number of confirmed cases as well as death cases of COVID-19 recorded in Nigeria for a period of 8 months.
Keywords: ARIMA, COVID-19, forecasting, time series