The ARIMA Procedure

Overview

The OUTLIER statement is a new enhancement to the ARIMA procedure. This statement is used as a diagnostic tool to detect shifts in the mean level of the response series that are not accounted for by the model in the preceding ESTIMATE statement. You can control different aspects of this detection process, such as the type and number of changes to detect, the significance level of the underlying tests, etc., by setting various options. Currently three types of changes in the mean level can be detected. These types are

Additive Outlier (AO), which represents a spurious change in one of the measurements
Level Shift (LS), which stands for a permanent shift in the level starting at some point in the observation period
Temporary Change (TC( [ d ] )), which represents a temporary change in the level of duration d

Chapter Contents
Previous
Next
Top