<![CDATA[The revenue of customs and excise is very important in APBN. By making accurate estimation, target of revenue can be better determined. In addition, the revenue of customs and excise is also influenced by many external factors that are difficult to predict therefore a rational approach is needed to estimate revenue. This research uses Double Exponential Smoothing, Ordinary Least Square (OLS) model and Moving Average in predicting customs and excise revenue. Data used in this research is secondary data in time coherent pattern. The data includes import duty, export duty and excise obtained from the Directorate General of Customs and excise (DJBC) in the form of annual and quarterly data. This data starts from 2002 to 2016 with out of sample from 2017 to 2019. Some of these models are compared to each other to obtain the best model, and from the best model is also obtained estimating results in 3 years ahead. This study shows that the Double Exponential Smoothing model is better for predicting import duties compared to OLS and Moving Average models, which are models that have the smallest Sum Square Error (SSE) value. While the export and excise duty is best estimated by using OLS model which is shown with coefficient of determination value (R2) regression model of export duty is 0.8, while the excise regression model has coefficient of determination of 0.9.Keywords: Customs Estimation, Double Exponential Smoothing, Ordinary Least Square, Moving Average]]>
Copyrights © 2018
