A novel hybrid ARIMA and regression tree model for the interval-valued time series

Abstract

With the arrival of the big data era, the interval-valued time series (ITS) has become a research hot spot. This study proposes a novel hybrid model combining the auto-regressive integrated moving average (ARIMA) and regression tree (RT) models for the ITS. Following the idea of hybrid ‘linear and nonlinear’ modeling framework, the ARIMA and RT models capture the linear and nonlinear components hidden in the ITS. The proposed ARIMA-RT model is compared with other competitors through a simulated experiment and a real ITS. Based on the experimental analysis, we find that the performance of the proposed ARIMA-RT model is strikingly superior to other competitors, notably in forecasting the nonlinear ITS. It indicates that the ARIMA-RT model has strong ability to capture the nonlinear ITS in stock markets.