Forecasting container throughput at Northport (Malaysia) Bhd using univariate modeling and box jenkins methodology / Nordina Amira Ramlan

The purpose of this study is to determine the most suitable technique between Univariate Modelling Technique and Box Jenkins Methodology Technique to generate forecast values for the total container throughputs using data taken from Northport (Malaysia) Bhd. The data was 10 years data and about 118 data. The models selected for Univariate Modelling Technique are Single Exponential, Double Exponential Smoothing, Holt’s Method, Adaptive Response Rate Exponential Smoothing and Holt-Winter’s Trend and Seasonality.These models are normally used to determine multiple-ahead-forecast, focus of one variable and large size of data. For Box jenkins Methodology, this study required to used SARIMA because there is existance of seasonality trend. The performance of the models were measured by calculating for the value of the Mean Square Error (MSE) and the Mean Absolute Percentage Error (MAPE). The best model is selected based on the smallest value of MSE and MAPE. In addition, the models were evaluated by comparing the forecast values generated by the models with the actual data. Based on the analysis, Holt-Winter’s Trend and Seasonality from Univariate Modeling is the most suitable model to forecast the monthly total container throughputs for Northport (Malaysia) Bhd for Univariate Modelling