Associate Professor

On Predicting Growth Factor of Daily New Cases Data of COVID-19 Epidemic in Spain Using ARIMA-ANN Hybrid Model
The Auto Regressive Integrated Moving Average (ARIMA) model cannot capture the nonlinear patterns exhibited by the 2019 coronavirus (COVID-19) in terms of daily growth factor of daily new cases data in Spain. As a result, Artificial Neural Networks (ANNs) model is commonly used to resolve problems with nonlinear estimation. Different models that include ARIMA, ANNs, seasonal decomposition of time series, and a combination of these three models (hybrid model) were proposed to forecast the Growth Factor of COVID-19. This study provides forecasting insights and criteria to use for similar time series data to predict growth factor of COVID-19 and selects the most suitable forecasting model for forecasting purpose. The best forecasting model selected was compared using the forecasting assessment criterion known as root mean squared error (RMSE) and mean absolute error (MAE). The results of this study add to the growing body of literature that seeks to accurately forecast the spread of COVID-19 by combining multiple models used by other researchers. The results are useful because it provides an accurate forecast for growth factor for COVID-19 epidemic. The study underscores the importance of appropriate forecasts for policy makers to enhance better decision making. All Governments and institutions involved in public health can benefit from these results for forecasting purposes using more reliable and accurate forecast model for COVID-2019 epidemic. The additional value of results is encouraging as the world struggles to restrain from the spread of COVID-19.
Keywords: COVID-19, Forecasting, Hybrid Model, ANN, ARIMA.