Documentation 
Evaluate standard errors for multivariate normal regression model
[StdParameters, StdCovariance] = mvnrstd(Data, Design, Covariance,
CovarFormat)
Data  NUMSAMPLESbyNUMSERIES matrix with NUMSAMPLES samples of a NUMSERIESdimensional random vector. If a data sample has missing values, represented as NaNs, the sample is ignored. (Use ecmmvnrmle to handle missing data.) 
Design  A matrix or a cell array that handles two model structures:

Covariance  NUMSERIESbyNUMSERIES matrix of estimates for the covariance of the regression residuals. 
CovarFormat  (Optional) String that specifies the format for the covariance matrix. The choices are:

[StdParameters, StdCovariance] = mvnrstd(Data, Design, Covariance, CovarFormat) evaluates standard errors for a multivariate normal regression model without missing data. The model has the form
$$Dat{a}_{k}\sim N\left(Desig{n}_{k}\times Parameters,\text{\hspace{0.17em}}Covariance\right)$$
for samples k = 1, ... , NUMSAMPLES.
mvnrstd computes two outputs:
StdParameters is a NUMPARAMSby1 column vector of standard errors for each element of Parameters, the vector of estimated model parameters.
StdCovariance is a NUMSERIESbyNUMSERIES matrix of standard errors for each element of Covariance, the matrix of estimated covariance parameters.
Note mvnrstd operates slowly when you calculate the standard errors associated with the covariance matrix Covariance. 
You can configure Design as a matrix if NUMSERIES = 1 or as a cell array if NUMSERIES ≥ 1.
If Design is a cell array and NUMSERIES = 1, each cell contains a NUMPARAMS row vector.
If Design is a cell array and NUMSERIES > 1, each cell contains a NUMSERIESbyNUMPARAMS matrix.
See Multivariate Normal Regression, LeastSquares Regression, CovarianceWeighted Least Squares, Feasible Generalized Least Squares, and Seemingly Unrelated Regression.
Roderick J. A. Little and Donald B. Rubin, Statistical Analysis with Missing Data, 2nd ed., John Wiley & Sons, Inc., 2002.