• Displays asymptotic correlation matrix of covariance parameter estimates. It is computed from the corresponding asymptotic covariance matrix (see the description of the ASYCOV option, below)
• This option requests that the asymptotic covariance matrix of the covariance parameters be displayed. By default, this matrix is the observed inverse Fisher information matrix, which equals 2 H -1, where H is the Hessian (second derivative) matrix of the objective function.
• Computes the estimated variance -covariance matrix of the fixed-effects parameters by using the asymptotically consistent (or sandwich) estimator.
• The GLIMMIX procedure normally computes various IC that typically apply a penalty to the (possibly restricted) log likelihood, log pseudo-likelihood, or log quasi-likelihood that depends on the number of parameters and/or the sample size .
• The GLIMMIX procedure normally computes various IC that typically apply a penalty to the (possibly restricted) log likelihood, log pseudo-likelihood, or log quasi-likelihood that depends on the number of parameters and/or the sample size.
• Select this option to request that the penalties include the number of fixed-effects parameters, when estimation in models with random effects is based on a residual (restricted) likelihood.Note : For METHOD=MSPL , METHOD=MMPL , METHOD=LAPLACE , and METHOD=QUAD , the IC=Q and IC=PQ options produce the same results.
• The GLIMMIX procedure normally computes various IC that typically apply a penalty to the (possibly restricted) log likelihood, log pseudo-likelihood, or log quasi-likelihood that depends on the number of parameters and/or the sample size.
• This is the default option for linear mixed model with normal errors, and the resulting information criteria are identical to the IC option specified using PROC MIXED Options .Note : For METHOD=MSPL , METHOD=MMPL , METHOD=LAPLACE , and METHOD=QUAD , the IC=Q and IC=PQ options produce the same results.
• Displays the parameter values at each iteration and enables the writing of notes to the SAS log pertaining to infinite likelihood and singularities during Newton-Raphson iterations.
• Note : This option was designed for use with analyses requiring extensive CPU resources.
• The RSPL option specifies that the estimation is based on a R esidual likelihood with a S ubject-specific expansion locus. The PL abbreviation identifies the method as a pseudo-likelihood technique.
• The MSPL option specifies that the estimation is based on a M aximum likelihood ( R ) with a S ubject-specific expansion locus. The PL abbreviation identifies the method as a pseudo-likelihood technique.
• The RMPL option specifies that the estimation is based on a R esidual likelihood with a M arginal-specific expansion locus. The PL abbreviation identifies the method as a pseudo-likelihood technique.
• The MMPL option specifies that the estimation is based on a M aximum likelihood with a M arginal-specific expansion locus. The PL abbreviation identifies the method as a pseudo-likelihood technique.
• Twice the negative of the resulting log -likelihood approximation is the objective function that the procedure minimizes to determine parameter estimates. Laplace estimates typically exhibit better asymptotic behavior and less small-sample bias than pseudo-likelihood estimators. On the other hand, the class of models for which a Laplace approximation of the marginal log likelihood is available is much smaller compared to the class of models to which PL estimation can be applied.
• Approximates the marginal log likelihood with an adaptive Gauss-Hermite quadrature.
• Compared to METHOD=LAPLACE , the models for which parameters can be estimated by quadrature are further restricted.
• For example, variance components have a default lower boundary constraint of 0, and the NOBOUND option allows their estimates to be negative.
• Requests that the starting values for the fixed effects not be obtained by first fitting a generalized linear model.
• Specifies that the levels of the classification variables are sorted in the order in which they appear in the input data set. Note : In generalized linear models with normally distributed data, you can use the PROFILE option to request profiling of the residual variance.
To specify more than one option, hold down as you left-click on the desired options.Refer to the SAS PROC GLIMMIX documentation for more information and references.