Variance Models

Variance model keywords

keywords are listed in three groups
  • Correlation matrices
  • Variance matrices
  • Fixed matrices

    Correlation matrices can be converted to variance matrices.

    The algebraic definitions of these structures are available in the User Guide.

    Correlation models.

    In the following table, w is the order of the matrix and the last column specifies the number of parameter values (initial values expected).
    Keyword Function Description Parameters
    Correlation matrices
    ID id() Identity 0
    AR1 ar1() First order autoregressive 1
    AR2 ar2() Autoregressive 2
    AR3 ar3() Autoregressive 3
    SAR sar() Symmetric autoregressive 1
    SAR2 sar2() Symmetric autoregressive 2
    MA, MA1 ma1() Moving Average 1
    MA2 ma2() Moving Average 2
    ARMA arma() Autoregressive Moving Average 2
    CORU coru() Uniform Correlation 1
    CORB corb() Banded Correlation w-1
    CORG corg() General Correlation w(w-1)/2
    EXP exp() Exponential 1
    GAU gau() Gaussian 1
    IEXP iexp() Isotropic Exponential 1
    IGAU igau() Isotropic Gaussian 1
    IEUC ieuc() Isotrophic Euclidean 1
    LVR lvr() Linear Variance 1
    SPH sph() Spherical 1
    CIR cir() Circular 1
    AEXP aexp() Anisotropic Exponential 2
    AGAU agau() Anisotropic Gaussian 2
    MATk matk() Matern k
    Variance structures
    DIAG diag() Diagonal w
    US us() Unstructured w(w+1)/2
    OWNk ownk() User supplied matrix k
    ANTEk antek() Antedependence w(w+1)/2
    CHOLk cholk() Cholesky - banded form w(w+1)/2
    CHOLkC cholkC() Cholesky - column form w(w+1)/2
    FAk fak() Factor Analytic (correlation form) w(k+1)
    FACVk facvk() Factor Analytic (covariance form) w(k+1)
    XFA k xfak() Extended Factor Analytic w(k+1)
    Fixed matrices
    AINV nrm() Numerator Relationship matrix (A) 0 or 1
    GIVk grmk() General (Inverse) Variance matrix 0 or 1

    See Also