A comparison of univariate methods for forecasting container throughput volumes

摘要: In this paper, six univariate forecasting models for the container throughput volumes in Taiwan's three major ports are presented. The include classical decomposition model, trigonometric regression model with seasonal dummy variables, grey hybrid and SARIMA model. purpose of paper is to search a that can provide most accurate prediction throughput. By applying monthly data these comparing results based on mean absolute error, percent error root squared we find general appears be best variations. result study may helpful predicting short-term variation demand other international ports.