Title: Modeling Data Containing Outliers using ARIMA Additive Outlier (ARIMA-AO)

Abstract: The aim this study is discussed on the detection and correction of data containing the additive outlier (AO) on the model ARIMA (p, d, q). The process of detection and correction of data using an iterative procedure popularized by Box, Jenkins, and Reinsel (1994). By using this method we obtained an ARIMA models were fit to the data containing AO, this model is added to the original model of ARIMA coefficients obtained from the iteration process using regression methods. This shows that there is an improvement of forecasting error rate data.