Modeling Data Containing Outliers using ARIMA Additive Outlier
\n(ARIMA-AO)

Abstract

The aim this study is discussed on the detection and correction of data containing the
\nadditive outlier (AO) on the model ARIMA (p, d, q). The process of detection and correction
\nof data using an iterative procedure popularized by Box, Jenkins, and Reinsel (1994). By using
\nthis method we obtained an ARIMA models were fit to the data containing AO, this model is
\nadded to the original model of ARIMA coefficients obtained from the iteration process using regression methods. In the simulation data is obtained that the data contained AO initial models are ARIMA (2,0,0) with MSE = 36,780, after the detection and correction of data obtained by the iteration of the model ARIMA (2,0,0) with the coefficients obtained from the regression Z Z Z X t X t X t t t t       0,106 0,204 0, 401 329 115 35,9   1 2 1 2 3       and MSE = 19,365. This shows that there is an improvement of forecasting error rate data.