MVTPDF Multivariate t probability density function (pdf).
    Y = MVTPDF(X,C,DF) returns the probability density of the multivariate t
    distribution with correlation parameters C and degrees of freedom DF,
    evaluated at each row of X.  Rows of the N-by-D matrix X correspond to
    observations or points, and columns correspond to variables or
    coordinates.  Y is an N-by-1 vector.

    C is a symmetric, positive definite, D-by-D matrix, typically a
    correlation matrix.  If its diagonal elements are not 1, MVTPDF scales
    C to correlation form.  DF is a scalar, or a vector with N elements.
  
    Example:

       C = [1 .4; .4 1]; df = 2;
       [X1,X2] = meshgrid(linspace(-2,2,25)', linspace(-2,2,25)');
       X = [X1(:) X2(:)];
       p = mvtpdf(X, C, df);
       surf(X1,X2,reshape(p,25,25));

    See also mvnpdf, mvtcdf, mvtrnd, tpdf.

    Reference page in Help browser
       doc mvtpdf

COPULAFIT Fit a parametric copula to data.
    RHOHAT = COPULAFIT('Gaussian', U) returns an estimate RHOHAT of the matrix
    of linear correlation parameters for a Gaussian copula, given data in U.  U
    is an N-by-P matrix of values in (0,1), representing N points in the
    P-dimensional unit hypercube.

    [RHOHAT, NUHAT] = COPULAFIT('t', U) returns an estimate RHOHAT of the matrix
    of linear correlation parameters for a t copula, and an estimate NUHAT of
    the degrees of freedom parameter, given data in U.  U is an N-by-P matrix of
    values in (0,1), representing N points in the P-dimensional unit hypercube.

    [RHOHAT, NUHAT, NUCI] = COPULAFIT('t', U) returns an approximate 95%
    confidence interval for the degrees of freedom parameter for a t copula,
    given data in U.

    PARAMHAT = COPULAFIT(FAMILY, U) returns an estimate PARAMHAT of the copula
    parameter for an Archimedean copula specified by FAMILY, given data in U.  U
    is an N-by-2 matrix of values in (0,1), representing N points in the unit
    square.  FAMILY is 'Clayton', 'Frank', or 'Gumbel'.

    [PARAMHAT, PARAMCI] = COPULAFIT(FAMILY, U) returns an approximate 95%
    confidence interval for the copula parameter from an Archimedean copula
    specified by FAMILY, given data in U.

    [...] = COPULAFIT(..., 'Alpha', ALPHA) returns an approximate 100(1-ALPHA)%
    confidence interval for the parameter estimate.

    COPULAFIT uses maximum likelihood to fit the copula to U.  When U contains
    data transformed to the unit hypercube by parametric estimates of their
    marginal cumulative distribution functions, this is known as the Inference
    Functions for Margins (IFM) method.  When U contains data transformed by
    the empirical CDF, this is known as Canonical Maximum Likelihood (CML).

    [...] = COPULAFIT('t', U, ..., 'Method', 'ApproximateML') fits a t copula by
    maximizing an objective function, as suggested by Bouye et al., that
    approximates the profile log-likelihood for the degrees of freedom parameter
    nu, for large sample sizes.  This method can be significantly faster than
    using maximum likelihood, however, it should be used with caution because
    the estimates and confidence limits may not be accurate for small or
    moderate sample sizes.  COPULAFIT('t', U, ..., 'Method', 'ML') is equivalent
    to the default maximum likelihood fit.

    [...] = COPULAFIT(..., 'Options', OPTIONS) specifies control parameters for
    the iterative algorithm used to compute estimates.  This argument can be
    created by a call to STATSET.  See STATSET('copulafit') for parameter names
    and default values.  This argument does not apply to the 'Gaussian' family.

    See also ecdf, copulacdf, copulapdf, copularnd.

    Reference page in Help browser
       doc copulafit

用法我知道啊，Matlab主页上也有，我想要相关函数！！就是相关M文件！！
谢谢