Enhancements in SAS/ETS^{®} 14.1 Software
Overview
SAS/ETS 14.1 introduces the X13 procedure and enhancements to many other procedures.
X13 Procedure
Because the US Census Bureau has included the X12ARIMA methodology in the X13ARIMASEATS program, the X12
procedure has been renamed the X13 procedure. You can continue to specify PROC X12 as before, but the X12ARIMA
methodology will be based on the current method in the X13ARIMASEATS method, including enhancements and
improvements.
The following features have been added to the X13 procedure:
 The DIFFID= option has been added to the AUTOMDL statement. The DIFFID= option specifies the estimation method
of identifying difference orders. This option is experimental in SAS/ETS 14.1.
 The DIFFIDITER= option has been added to the AUTOMDL statement. The DIFFIDITER= option specifies the maximum number of iterations for exact
likelihood for DIFFID=EXACTFIRST. This option is experimental in SAS/ETS 14.1.
 The EXACT= option has been added to the ESTIMATE statement. The EXACT= option specifies the likelihood function
to be used for estimating autoregressive (AR) and moving average (MA) parameters. This option is experimental in SAS/ETS 14.1.
 The EASTERMEANS= option has been added to the REGRESSION statement. The EASTERMEANS= option specifies the method
to be used to calculate the means for the Easter regression variable.
 The following tables have been added to the displayed output for automatic modeling: "Check of the
Residual LjungBox Q Statistic," "Initial Automatic Model Selection," and "Final Checks for Identified
Model." In addition, if the orders have been altered during the final checks, the new Orders Altered column
in the "Final Automatic Model Selection" table displays a value of Yes. These changes clarify the final
automatic model selection process.
 The default for the MAXITER= option in the ESTIMATE statement has been changed to 1,500.
 The following tables have been added to the tables available in the X11 statement: B7, B13, B17, B20, C1, D8B, and E18.
COUNTREG Procedure
The following features have been added to the COUNTREG procedure:
 The following Bayesian analysis features have been added:
 harmonic mean evaluation of the marginal likelihood to compare competing models (this estimator
does not require additional simulations after the posterior samples have been obtained)
 evaluation of the marginal likelihood using an importance sampling algorithm that is based on the
crossentropy theory (this estimator requires additional importance sampling simulations after the
posterior samples have been obtained)
 The TEST statement has been added.
 Spatial lag models have been added, enabling you to include spatial effects in a model. The
SPATIALEFFECTS, SPATIALDISPEFFECTS, and SPATIALZEROEFFECTS statements have been added to allow for spatial
effects in the MODEL, DISPMODEL, and ZEROMODEL statements, respectively. In addition, variable selection
functionality has been added to the DISPMODEL, SPATIALEFFECTS, SPATIALDISPEFFECTS, and SPATIALZEROEFFECTS
statements.
HPCOUNTREG Procedure
The following features have been added to the HPCOUNTREG procedure:
 TEST statement
 support for the ConwayMaxwell Poisson distribution
HPPANEL Procedure
The following features have been added to the HPPANEL procedure:
 the betweengroups estimator, which is equivalent to regression on crosssectional means (to obtain this
estimator, specify the BTWNG option in the MODEL statement)
 the betweentimeperiods estimator, which is equivalent to regression on means at each time point (to
obtain this estimator, specify the BTWNT option in the MODEL statement)
 pooled OLS regression, which is obtained by specifying the POOLED option in the MODEL statement
MODEL Procedure
The %EQAR and %EQMA macros have been added to the MODEL procedure. Model programs that are expressed using
general form equations can now use the %EQAR macro to specify autoregressive error processes or the %EQMA macro
to specify moving average error processes. The %EQAR and %EQMA macros modify general form equations similar to
how the %AR and %MA macros modify normal form equations.
PANEL Procedure
The following features have been added to the PANEL procedure:
 The Hausman and Taylor and Amemiya and MaCurdy estimators, which are hybrids that combine the desirable properties of
fixedeffects and randomeffects models. Under the right circumstances, these estimators afford you the
consistency of fixed effects and the efficiency and wider applicability of random effects. Both estimators
are instrumentalvariables regressions, where you stipulate a set of regressors as correlated with
individual effects. The instrumental variables are then determined internally from the set of uncorrelated
regressors, their individuallevel means, and their deviations from individuallevel means. To obtain these
estimators, first specify the correlated regressors in the CORRELATED= option in the INSTRUMENTS statement,
and then specify the HTAYLOR or AMACURDY option in the MODEL statement.
 Comparison tables for multiple models. The new COMPARE statement creates tables of sidebyside
comparisons of parameter estimates and other model statistics. You can fit multiple models in the PANEL
procedure by issuing multiple MODEL statements. Also specifying a COMPARE statement creates tables that
compare the models. The COMPARE statement creates two tables: the first table compares model fit statistics
such as R^{2} and MSE; the second table compares regression coefficients, their
standard errors, and (optionally) t tests.
 More general Hausman specification tests. In previous versions, Hausman tests for random effects
required that the randomeffects model contain no timeinvariant regressors (regressors that would be
dropped from the fixedeffects model). That requirement has been relaxed in SAS/ETS 14.1, and the Hausman
test is now a comparison of regressors that are common to both the random and fixedeffects models. A new
column labeled "Coefficients" has been added to the output table for the Hausman test. The "Coefficients"
column tells you how many coefficients are common to both models, and thus also tells you the nominal rank
of the test.
QLIM Procedure
The following features have been added to the QLIM procedure:
SASECRSP Interface Engine
If you install SAS/ETS on a Windows system, you no longer need to install the CRSPAccess API, because it is
now distributed automatically during the installation. Before you run SASECRSP, your Windows setup requires
that the CRSPDB_SASCAL environment variable be set to the path where your database calendar files reside.
SASEFRED Interface Engine
The following features have been added to the SASEFRED interface engine:
 Linux X64 (LAX) host support has been added.
 The PROXY= option specifies a proxy server and port number to use if the connection times out without a
proxy server.
 The RTSTART= and RTEND= options support the realtime periods for FRED data. Because the default is
today, it is important to support the range of first to last available (the complete realtime period), in
addition to other ranges for which data were available.
 You can use the URL= option to request the following useful information about categories, tags, groups,
and releases:
 a list of the vintage dates (release dates) for a particular series that is specified in the
IDLIST= option<
 a list of the available series for a particular release or for a specified source, tag name, or
category ID
 a list of available sources for today's date
 a list of the categories available for a specified series ID
 Blanks are now allowed in pathnames that are used in SASEFRED options.
 The DEBUG=ON option logs diagnostics in the SAS log.
SASEXCCM Interface Engine
If you install SAS/ETS on a Windows system, you no longer need to install the CRSPAccess API, because it is
now distributed automatically during the installation. Your Windows setup does not require any special
environment variables.
SASEXFSD Interface Engine
The following features have been added to the SASEXFSD interface engine:
 Linux X64 (LAX) host support has been added.
 The PROXY= option specifies a proxy server and port number to use if the connection times out without a
proxy server.
 The CONNECT= option specifies whether to connect using the secure HTTP address in the PROXY=
specification to obtain a secure connection.
 Blanks are now allowed in pathnames that are used in SASEXFSD options.
 The UNIVERSE= option is now supported on the ExtractFormulaHistory factlet.
 The DEBUG=ON option logs diagnostics in the SAS log.
SASEQUAN Interface Engine
The following features have been added to the SASEQUAN interface engine:
 Linux X64 (LAX) host support has been added.
 The PROXY= option specifies a proxy server and port number to use if the connection times out without a
proxy server.
 When a long variable name (more than 32 bytes) is truncated, the 32byte name might not be unique, so
SASEQUAN appends the variable number to the name in order to create a unique variable name.
 Blanks are now allowed in pathnames that are used in SASEQUAN options.
 Up to nine Quandl codes are allowed in the IDLIST= option. SASEQUAN returns an error message when more
than nine are specified.
 The DEBUG=ON option logs diagnostics in the SAS log.
SSM Procedure
The DEPLAG statement has been added to the SSM procedure. It simplifies the specification of models that have
lagged values of response variables in the observation equation. The DEPLAG statement enables you to define a
linear combination of lagged response variables, which can be subsequently used as a righthandside term in the
MODEL statement. Models that include lagged response variables are permitted only if the data form a time series
(either univariate or multivariate).
VARMAX Procedure
The following features have been added to the VARMAX procedure:
 Vector error correction models in ARMAGARCH form are supported. You can use the COINTEG statement
together with the Q= option in the MODEL statement and the GARCH statement to model the cointegration
relationship between multiple time series that have GARCHtype innovations.
 The linear equality and inequality constraints for any parameters to be estimated in vector error
correction models are supported. You can use the BOUND and RESTRICT statement to study the restricted
cointegrated systems.
 The covariance and standard errors of the parameter estimates of the adjustment coefficient matrix and
the covariance matrix of innovations in vector error correction models are supported. The outputs of
parameter estimates of the longrun parameters and the error correction trend parameters are also
supported.
 You can apply the Wald tests, by using the TEST statement, on any parameters in vector correction models
except the longrun parameters and the error correction trend parameters.
 You can specify initial values, by using the INITIAL statement, for any parameters to be estimated in
vector error correction models. If you do so, you must specify the fullrank initial matrices for both the
adjustment coefficient matrix and the longrun parameters.

A new estimation method, the conditional maximum likelihood method (CML), is supported. This method is
especially suitable for estimating VARMAX models on large samples.

The log likelihoods for all types of models are output. These outputs are especially useful if you need
to execute the likelihood ratio (LR) tests.
 Definitions have been revised for the following information criteria: Akaike's information criterion (AIC), the
corrected Akaike's information criterion (AICC), the HannanQuinn criterion (HQC), and the Schwarz Bayesian
criterion (SBC, also referred to as BIC). You can compare more types of models, including all forms of
multivariate GARCH models.
 The ECTREND option is supported in the COINTEG statement. All options in the ECM= option in the MODEL
statement are now supported in the COINTEG statement, and the ECM= option becomes obsolete. Starting with
SAS/ETS 14.1, it is recommended that you use the COINTEG statement instead of the ECM= option to fit vector
error correction models.
 The new NLC option, is supported in the COINTEG statement. This option enables you to
explicitly require that the adjustment coefficient matrix and the longrun parameters both be full rank
when you numerically maximize the likelihood of a vector error correction model.