Application of a Mixed Regression Model

The forecast of volumes of import and export containers is one of the most important issues for government transportation departments and seaport organizations. Although a number of studies focus on the subject of forecast of import and export containers in the past, most of these studies do not consider the forecast error due to the “non-stationary”relationship between the volumes of import and export containers and the economic variables. Thus this article attempts to fill this gap in the current literature by establishing a modified non-linear regression model. An empirical study in Taiwan is conduced to demonstrate the effectiveness of the proposed modified non-linear regression model. Finally, this paper compares the accuracy of this modified non-linear regression model, the traditional linear regression model and the traditional non-linear regression model for forecasting the volumes of Taiwan’s import containers. The comparison results show that the proposed modified non-linear regression model in this paper exhibits higher prediction accuracy than previous models.