GAUSS Procedures This page contains (will contain) various GAUSS procedures to adaptively estimate a variety
of time series and other models.  The following is the list of procedures available along with a
brief description of the contents.  The setup and maintenance of this page is funded by NSF
grant #SBR-970159.  These programs are for public noncommercial use. We make no
performance guarantees. Primary Authors:  Douglas Hodgson, Keith Vorkink, and Irina Solyanik.
Most Recent Update:  January, 2002. Programs
  ? ADECM     Included are the procedures adecm.g and jmle.g along with a file
                        adecm.rdme which discusses installation and the estimation procedure.  The
                        two .g files adaptively estimate the cointegrating matrix and error correction
                        matrix in an Error Correction Model. This procedure implements the estimator
                        discussed in the paper:  Hodgson, D. (95), "Adaptive Estimation of Error
                        Correction Models," Econometric Theory 14, 44-69. DOWNLOAD
  
  ? ADRMA
    Included are the procedure adrma.g along with a file adrma.rdme
                        which discusses installation and the estimation procedure.  Adrma.g procedure
                        adaptively estimates the cointegrating matrix in the estimation of a cointegrating
                        regression where the residuals are allowed to follow an ARMA process. This
                        procedure will implement the estimator discussed in the paper:  Hodgson, D.,
                        "Adaptive Estimation of Cointegrating Regressions with ARMA Errors,"
                        Forthcoming in Journal of Econometrics(98).  DOWNLOAD
  
  ? STEIG
       Included are the procedure steig.g along with a file steig.rdme which
                        discusses installation and the estimation procedure.  Steig.g procedure
                        adaptively estimates the coefficient vector in the estimation of a linear regression
                        model where the residuals are allowed to follow an ARMA process. This
                        procedure will implement the estimator discussed in the paper:  Steigerwald, D.,(92)
                        "Adaptive Estimation in Time Series Regression Models," Journal of
                        Econometrics 54, 251-275.  We note this estimation procedure generalizes
                        two well know models.  When no serial dependence exists in the residuals the
                        model reduces to Bickel's(82, Annals of Statistics) model.  When no
                        regressors are present the model reduces to Kreiss' (87, Annals of Statistics)
                        model.  DOWNLOAD
  
  ? ADSUR
       Included are the procedure adsur.g with a file adsur.rdme which
                        discusses installation and the estimation procedure. Adsur.g adaptively
                        estimates the coefficient vector in the estimation of a Seemingly Unrelated
                        Regression Model (SUR). The procedure provides estimates of the system using
                        GLS, one-step adaptive, and an iterative adaptive estimator. The procedures
                        implement the estimator discussed in Hodgson, D., O. Linton, and K. Vorkink
                        (2001), "Testing the Capital Asset Pricing Model Efficiently Under Elliptical
                        Symmetry:  A Semiparametric Approach" forthcoming in Journal of Applied
                        Econometrics.   DOWNLOAD
  ? ADCAPM
  Included are the procedure adcapm.g with a file adcapm.rdme which
                        discusses installation and the estimation procedure. Adcapm.g adaptively estimates the
                        coefficient vector in the estimation of a linear regression model . The procedure provides
                        estimates of the parameters using OLS, one-step adaptive, and an iterative adaptive
                        estimates.  The procedures implement the estiamator discussed in K. Vorkink(2001),
                        "Return Distributions and Improved Tests of Asset Pricing Models," working Paper,
                         Marriott School of Management.  DOWNLOAD
  
  ? WALD        Included are the procedure wald.g along with a file wald.rdme which
                        discusses installation and the estimation procedure.  Wald.g procedure
                        constructs and performs wald tests.  The restrictions must be linear and both
                        parameter estimates and covariance matrix of parameters are required inputs
                        for the procedure.  This procedure can be used in conjunction with the above
                        estimation procedures. DOWNLOAD
  
  ? DEN
         Included are the procedures den.g and syden.g along with a file
                        density.rdme which discusses installation and the estimation procedure.
                        Den.g nonparametrically estimates the density of a nxm zero mean series
                        (m ?=2).  Syden.g nonparametrically estimates the density of an nxm series
                       assuming symmetry. DOWNLOAD
  
  ? ADARCH
       Included are the procedure adarch.g along with a file adarch.rdme which
                        discusses installation and the estimation procedure . The procedure implements
                        adaptive estimation of parameters of ARCH model discussed in O. Linton(93),
                         "Adaptive Estimation in ARCH Models", Econometric Theory, 9, pp.539-569.
                        Procedure keeps the overall scale parameter of ARCH model fixed and computes
                        adaptive estimates of regression parameters and identifiable ARCH parameters.
                        The error density is assumed to be symmetric about zero. DOWNLOAD
  
  ? ADUNIT
       Included are the procedure adunit.g, adunit.rdme which discusses installation
                        and the estimation procedure, and adunit.tex which also discussed the estimation and
                        testing procedure. The procedure implements adaptive unit root tests discussed in
                        O. Beelders(98), "Adaptive Unit Root Tests", Working Paper, Emory University.
                        The error density is assumed to be symmetric about zero.  DOWNLOAD
  
  ? ADAR
       Included are the procedure adar.g, adar.rdme which discusses installation
                        and the estimation procedure Procedure adar.g is the main  procedure to call
                        to estimate parameters of AR(p) model. Procedure constructs an estimator which
                        is adaptive for all densities of the distribution of the white noise.The main reference
                        is Jens-Peter Kreiss(1987), "On Adaptive Estimation In Autoregressive Models
                       When There Are Nuisance Functions", Statistics & Decisions, 5, pp. 59-76. DOWNLOAD
  
  
  ? SEMIPARMA
       Included are the procedure semiparma.g, semiparma.rdme which discusses
                        installation and the estimation procedure. Procedure semiparma.g is the main
                        procedure to call to estimate parameters of linear regression model with ARMA
                        errors. The main reference is Douglas Hodgson (1998), "Semiparametric Efficient
                        Estimation in Time Series Regression", University of Rochester Manuscript.
                        Procedure computes semiparametric estimates of parameters of linear regression
                        model with ARMA errors in which the innovations to the ARMA process is
                        stationary and ergodic martingale difference sequence that is also 1st order Markov
                        process. Conditional density g(e(t)|e(t-1)) is assumed to be symmetric. The
                        semiparametric efficiency bound is also reported.  DOWNLOAD
  
  ? ADTEST
       Included are the procedure adtest.g along with a file adtest.rdm, which discusses
                        installation and the estimation procedure . The procedure constructs
                        a test statistic for specification tests of conditional heteroscedasticity. Theoretical
                        details are discussed in Linton, O., and D. Steigerwald (2000) "Adaptive Testing
                        in ARCH Models", Econometric Reviews, 19(2): 146-174.
                        The semiparametric test statistic is constructed from a nonparametric estimator
                        of the innovation density and is adaptive (i.e., asymptotically equivalent to the test
                        statistic constructed from the true likelihood). The test statistic maximizes
                        asymptotic local power and weighted average power criteria for the general
                        family of densities. The asymptotic distribution of test statistic is Gaussian under
                        the sequence of local alternatives and is standard Gaussian under the null.
                        Procedure also constructs the semiparametric estimator of the full parameter
                        vector and its asymptotic covariance. DOWNLOAD
  
  
  *If you need WinZip to extract the above programs, follow the link below to obtain a copy of the program.
WinZip
  Return to Douglas Hodgson's Page Return to Keith Vorkink's Page