The X12 Procedure

SEATSDECOMP OUT= Data Set

You can use the SEATSDECOMP statement to output the component series that is computed using the SEATS method of seasonal decomposition.

The OUT= data set specified in the SEATSDECOMP statement contains the BY variables (if any), the ID variables (if any), and either the DATE= variable if the DATE= option is specified or the variable _DATE_ if the DATE= option is not specified. If user-defined regressor variables or EVENT variables are specified, they are included. In addition, the five components computed by the SEATS decomposition method are included in the OUT= data set for each series.

The SEATSDECOMP OUT= data set can contain the following variables:

BY variables

are the BY variables used to subset the series by BY groups. The BY variables included in this data set match the BY variables (if any) that are used to process the series in the DATA= data set.

ID variables

enable the series observations to be identified using further information. The ID variables included in this data set match the ID variables (if any) that are specified in the ID statement and input from the DATA= data set.

DATE variable

is the time ID variable used to process the time series. It is either the variable specified in the DATE= option in the PROC X12 statement or the variable _DATE_ that is generated by the START= option in the PROC X12 statement.

_YEAR_ variable

contains a value for the year of the date variable for the observation. This variable is included in the OUT= data set if YEARSEAS is specified in the OUTPUT statement.

_SEASON_ variable

contains a value for the month or quarter of the date variable for the observation. This variable is included in the OUT= data set if YEARSEAS is specified in the OUTPUT statement.

User-defined variables

are variables specified in the INPUT statement or the USERVAR= option in the REGRESSION statement. The values of these variables are copied from the DATA= data set or from the AUXDATA= data set.

EVENT variables

are variables that are specified in the EVENT statement. The values of these are computed based on the event definition and the dates of the time series observations.

Component variables

contains the data from the SEATS decomposition tables. The variable name used in the output data set is the input variable name followed by an underscore and the corresponding table name.

<variable>_OS

contains the original series for SEATS decomposition. This is the B1 series from the X-12-ARIMA method.

<variable>_SC

contains the seasonal component series that is calculated by SEATS decomposition.

<variable>_TC

contains the trend component series that is calculated by SEATS decomposition.

<variable>_SA

contains the seasonally adjusted series that is calculated by SEATS decomposition.

<variable>_IC

contains the irregular series that is calculated by SEATS decomposition.