Modeling with Auto-Regressive Integrated Moving Average and Artificial Neural Network Hybrid Model in predicting long term Healthcare cost in Rwanda

Abstract

The ARIMA model has been the preferred model for time series forecast for decades. Moreover,two decades ago, the artificial neural networks (ANNs) models have got attention to model the phe nomena with a complexity of relationships. However, many of existing researches have identified a mixture of results with the performance of the ANN model compared to the ARIMA model. In this sense, we propose a hybrid model, which differs in combining the advantages of ARIMA and ANNs in capturing the linear and non-linear relationship in data. The ARIMA-ANNs model was tested on sets of health expenditure actual data. Our results showed the effectiveness of the hybrid model which the higher accuracy in predicting as compared to existing models. In view of the fact, the empirical results from all from three models considered in this study showed that from 2006 looking forward to 2027, there will be an increasing trend of health expenditure. The results showed that the forecasted values by ARIMA models will be 557,299.97 ; 37,998.01 and 519,632.64 million Rwandan francs for Total health expenditure, Government health expenditure and household health expenditure respectively in 2027. While using the ANNs models, the forecasted values are 555,090.65 ; 37,847.37 ;517,572.65 million Rwandan francs for Total health expenditure, Government health expenditure and household health expenditure respectively in 2027 And, ARIMA-ANN models forecasted values in 2027 for Total health expenditure, Government health expenditure and household health expenditure are 552,881.33 ; 37,696.74 ; 515,512.66 in million Rwandan respectively. This study recommends the use of ARIMA ANN hybrid while modeling the health expenditure.