Modeling and Forecasting of US Health Expenditures Using ARIMA Models

Abstract

This paper presents the practical steps to be analyzed in order to use autoregressive integrated moving average (ARIMA) time series models to forecast the total health expenditures, as a percentage of GDP, for the USA. The aim of this study is to identify the appropriate type of model based on the Box–Jenkins methodology. In particular, we apply the static one-step ahead forecasting method to the annual data over the period 1970–2015. The results from this study show that ARIMA (0,1,1) model is the appropriate model to forecast the US health expenditures in this period.