The X13 Procedure

Functional Summary

Table 45.1 summarizes the statements and options that control the X13 procedure.

Table 45.1: X13 Syntax Summary

Description

Statement

Option

Data Set Options

   

Specifies the auxiliary data set

PROC X13

AUXDATA=

Specifies the input data set

PROC X13

DATA=

Specifies the user-defined event definition data set

PROC X13

INEVENT=

Specifies regression and ARIMA information

PROC X13

MDLINFOIN=

Outputs regression and ARIMA information

PROC X13

MDLINFOOUT=

Writes summary statistics to an output data set

PROC X13

OUTSTAT=

Writes table values to an output data set

OUTPUT

OUT=

Appends forecasts to the OUTPUT OUT= data

X11 or

OUTFORECAST

set

FORECAST

 

Prefixes backcasts to the OUTPUT OUT= data set

FORECAST

OUTBACKCAST

Display Control Options

   

Suppresses all displayed output

PROC X13

NOPRINT

Specifies the plots to be displayed

PROC X13

PLOTS=

Specifies the type of spectral plot to be

PROC X13

PERIODOGRAM

displayed

   

Specifies the series for spectral analysis

PROC X13

SPECTRUMSERIES=

Displays automatic model information

AUTOMDL

PRINT=

Specifies the number of lags in regARIMA

CHECK

MAXLAG=

model residuals ACF and PACF tables and plots

   

Displays regARIMA model residuals

CHECK

PRINT=

information

   

Displays the iterations history

ESTIMATE

ITPRINT

Displays information about restarted iterations

ESTIMATE

PRINTERR

Specifies the differencing used in the ARIMA model identification ACF and PACF tables and plots

IDENTIFY

DIFF=

Specifies the seasonal differencing used in the ARIMA model identification ACF and PACF tables and plots

IDENTIFY

SDIFF=

Specifies the number of lags in ARIMA model

IDENTIFY

MAXLAG=

identification ACF and PACF tables and plots

   

Displays regression model parameter estimates

IDENTIFY

PRINTREG

Requests tables that are not displayed by

TABLES

 

default

   

Specifies that the summary line not be

TABLES

NOSUM

displayed

   

Date Information Options

   

Specifies the date variable

PROC X13

DATE=

Specifies the date of the first observation

PROC X13

START=

Specifies the beginning or ending date or both of the subset

PROC X13

SPAN=

Specifies the interval of the time series

PROC X13

INTERVAL=

Specifies the interval of the time series

PROC X13

SEASONS=

Declaring the Role of Variables

   

Specifies BY-group processing

BY

 

Specifies identifying variables

ID

 

Specifies the variables to be seasonally adjusted

VAR

 

Specifies the user-defined variables that are

USERDEFINED

 

available for regression

   

Controlling the Table Computations

   

Suppresses trimming of leading and trailing missing values (if they exist)

PROC X13

NOTRIMMISS

Transforms or prior-adjusts the series

TRANSFORM

FUNCTION=

Transforms or prior-adjusts the series

TRANSFORM

POWER=

Adjusts the series by using a predefined

ADJUST

PREDEFINED=

adjustment variable

   

Specifies the likelihood function to be used for estimating AR and MA parameters

ESTIMATE

EXACT=

Specifies the maximum number of iterations for estimating AR and MA parameters

ESTIMATE

MAXITER

Specifies the convergence tolerance for

ESTIMATE

TOL=

nonlinear estimation

   

Specifies size of forecast confidence limits

FORECAST

ALPHA=

Specifies the number of backcasts by which to extend the series for seasonal adjustment

FORECAST

NBACKCAST=

Specifies the number of forecasts by which to extend the series for seasonal adjustment

FORECAST

LEAD=

Specifies that one-step-ahead forecasts be computed

FORECAST

OUT1STEP

Specifying Outlier Detection Options

   

Specifies automatic outlier detection

OUTLIER

 

Specifies the span for outlier detection

OUTLIER

SPAN=

Specifies the outlier types to be detected

OUTLIER

TYPE=

Specifies the critical values for outlier detection

OUTLIER

CV=

Specifies the critical values for AO outlier

OUTLIER

AOCV=

detection

   

Specifies the critical values for LS outlier

OUTLIER

LSCV=

detection

   

Specifies the critical values for TC outlier

OUTLIER

TCCV=

detection

   

Specifies the alpha value for outlier detection

OUTLIER

ALPHA=

Specifies the method for calculating the critical

OUTLIER

CVMETHOD=

value for outlier detection based on the alpha value

   

Specifies the number of level-shift outliers to

OUTLIER

LSRUN=

consider for forming a temporary level-shift

   

Specifies the rate of decay for temporary change outliers

OUTLIER

TCRATE=

Specifies the method of adding outliers at each iteration

OUTLIER

METHOD=

Specifies the difference in critical values

OUTLIER

ALMOST=

for almost outliers

   

Specifying the Regression Model

   

Specifies regression variables to be

REGRESSION

AICTEST=

selected using an AIC-based test

   

Specifies predefined regression variables

REGRESSION

PREDEFINED=

Specifies user-defined regression variables

REGRESSION

USERVAR=

Specifies user-defined regression variables

INPUT

 

Specifies user defined event regression

EVENT

 

variables

   

Specifies the method used to calculate the

REGRESSION

EASTERMEANS=

means for the Easter regression variable

   

Specifies which types of regression effects

REGRESSION

NOAPPLY=

are not to be removed before seasonal

   

adjustment

   

Specifying the ARIMA Model

   

Uses the X-13ARIMA-SEATS TRAMO-based

AUTOMDL

 

method to choose a model

   

Chooses a regARIMA model

PICKMDL

 

from a set that you specify

   

Specifies the ARIMA part of the model

ARIMA

MODEL=

Specifying Automatic Model Detection

   

Options

   

Specifies the maximum orders of ARMA

AUTOMDL

MAXORDER=

polynomials

   

Specifies the maximum orders of differencing

AUTOMDL

MAXDIFF=

Specifies the estimation method for identifying difference orders

AUTOMDL

DIFFID=

Specifies the maximum number of iterations for exact likelihood for DIFFID=EXACTFIRST

AUTOMDL

DIFFIDITER=

Specifies the fixed orders of differencing

AUTOMDL

DIFFORDER=

Suppresses fitting of a constant parameter

AUTOMDL

NOINT

Specifies the preference for balanced models

AUTOMDL

BALANCED

Specifies Hannan-Rissanen initial estimation

AUTOMDL

HRINITIAL

Specifies default model acceptance based on Ljung-Box Q

AUTOMDL

ACCEPTDEFAULT

Specifies the acceptance value for Ljung-Box Q

AUTOMDL

LJUNGBOXLIMIT=

Specifies the percentage by which to

AUTOMDL

REDUCECV=

reduce the outlier critical value

   

Specifies the critical value for ARMA

AUTOMDL

ARMACV=

coefficients

   

Model Diagnostics

   

Examines the regARIMA model residuals

CHECK

 

Specifying Seasonal Adjustment Options

   

Specifies seasonal adjustment

X11

 

Specifies the mode of seasonal adjustment

X11

MODE=

decomposition

   

Specifies the seasonal filter

X11

SEASONALMA=

Specifies the sigma limits

X11

SIGMALIM=

Specifies the Henderson trend filter

X11

TRENDMA=

Specifies the D11 calculation method

X11

TYPE=

Specifies the adjustment factors to remove

X11

FINAL=

from final seasonally adjusted series

   

Specifies a method for reconciling the

X11

FORCE=

seasonally adjusted series to the

   

original series

   

Specifies that SEATS seasonal

SEATSDECOMP

OUT=

decomposition be output to a data set