Adding Exogenous Variable in Forming ARIMAX Model to Predict Export Load Goods in Tanjung Priok Port

Abstract

The main idea of world maritime has been launched by the Indonesia’s Government through the development of inter-island connectivity, namely a logistics distribution line system using cargo ships with scheduled routes. However, in terms of inter-island sea transportation connectivity using sea transportation, the number of ships used for loading and unloading activities at Tanjung Priok in 2020 reached 11,876 units, which number decreased by 12.6% compared to the previous year, this figure was not sufficient for transportation of Indonesian loading and unloading goods (exports). This condition is important to note because the implementation of sea transportation, especially for sea toll transportation, if it cannot reach all regions, will cause freight transportation in some areas to be limited and regional economic growth cannot be distributed evenly. The purpose of this study is to predict the number of goods loaded (exported) at the Port of Tanjung Priok, by establishing an export forecasting model. Exogenous variable in the form of the Indonesian Wholesale Price Index. After analyzing the data, the order of the ARIMA model (5,1,1) was obtained as a parameter to estimate the ARIMAX model. From the ARIMAX model (5,1,1), the model's accuracy rate is 13.25% which is quite feasible to use to predict the total export cargo for the period January 2021-December 2021. Forecasting results show better fluctuations than in 2020.

The Effect of Ensemble Averaging Method on Rainfall Forecasting in Jakarta Using ARIMA and ARIMAX

“…Furthermore, using the ARIMA model that has been significant in the previous stage, a parameter significance test is carried out for the ARIMAX model with the addition of exogenous variables (Elvina et al, 2023). Models with all significant model parameters are continued to the diagnostic test stage of the ARIMAX model (Ahmar et al, 2022).…”

Forecasting Stock Price Using Armax-Garchx Model During The Covid-19 Pandemic

“…Several researchers who have forecasted using the ARMAX model include (Pradana et al, 2022), (Riestiansyah et al, 2022) and (Elvina et al, 2023). The results of their research show that the ARMAX model is more accurate in forecasting than the ARMA model.…”