AUTOOUTLIER Subcommand (TSMODEL command)
The AUTOOUTLIER
subcommand
specifies whether outlier time points are identified automatically.
- By default, outliers are not identified.
- If identified, outliers are not discarded but are incorporated in the model.
- The
AUTOOUTLIER
subcommand is ignored with a warning if it is part of aMODEL
block that contains anEXSMOOTH
subcommand.
Example
TSMODEL
/MODELDETAILS PRINT=PARAMETERS
/MODEL DEPENDENT=revenue
/EXPERTMODELER TYPE=[ARIMA]
/AUTOOUTLIER DETECT=ON.
DETECT Keyword
The DETECT
keyword controls
whether automatic outlier detection is performed.
OFF. Outliers are not identified. This setting is the default.
ON. Outliers are
identified.
ON
is
ignored with a warning if the model type is EXSMOOTH
.
TYPE Keyword
The optional TYPE
keyword
specifies the types of outliers that are identified when DETECT=ON
. TYPE
is ignored if DETECT=OFF
.
- Additive and level shift outliers are identified by default.
- The specification of outlier types overrides the default. That is, if you specify a list of outlier types, additive and level shift outliers are not included unless explicitly requested.
The keyword is followed by an equals sign (=) and one or more of the following options enclosed in square brackets:
ADDITIVE. Additive. An outlier that affects a single observation. For example, a data coding error might be
identified as an additive outlier. Default if DETECT=ON
.
LEVELSHIFT. Level shift. An outlier that shifts all observations by a constant, starting at a particular series
point. A level shift could result from a change in policy. Default if DETECT=ON
.
INNOVATIONAL. Innovational. An outlier that acts as an addition to the noise term at a particular series point. For stationary series, an innovational outlier affects several observations. For nonstationary series, it may affect every observation starting at a particular series point.
TRANSIENT. Transient. An outlier whose impact decays exponentially to 0.
SEASONALADDITIVE. Seasonal additive. An outlier that affects a particular observation and all subsequent observations
separated from it by one or more seasonal periods. All such observations are affected equally. A
seasonal additive outlier might occur if, beginning in a certain year, sales are higher every
January. The keyword generates an error if the season
length has not been defined via the DATE
command or the SEASONLENGTH
keyword.
LOCALTREND. Local trend. An outlier that starts a local trend at a particular series point.
ADDITIVEPATCH. Additive patch. A group of two or more consecutive additive outliers. Selecting this outlier type results in the detection of individual additive outliers in addition to groups of them.