GLMM Save Options

Use this to save results from a GLMM analysis in Genstat data structures.

After selecting the appropriate boxes, type the identifiers of the data structures into the corresponding In: fields.

Save

Select the results you want to save.

Residuals	Variate	Residuals from the fitted model using the specified Terms for residuals
Fitted values	Variate	Fitted values from the model
Dispersion	Scalar	The estimated dispersion
Components	Variate	Estimated variance components
Variance-covariance matrix for components	Symmetric matrix	Variance-covariance matrix of the variance components
Variance-covariance matrix for means	Symmetric matrix	Variance-covariance matrix of the means
Variance parameter var-cov matrix	Symmetric matrix	Variance-covariance matrix of the variance parameters
Variance parameter labels	Text	The descriptions of variance parameters
Estimates of missing values	Variate	The estimates for the missing values
SE of missing value estimates	Variate	The standard errors for the estimates for the missing values
Units of missing values	Variate	The unit numbers of the missing values. If you wanted to replace the missing values with their estimates, you would use CALC Y$[units] = mv_est
Deviance	Scalar	Deviance from the model – the type is specified by the Deviance to use option
Fixed degrees of freedom	Scalar	The number of degrees of freedom for the fixed model
Random degrees of freedom	Scalar	The number of degrees of freedom for the random model
Iterative weights	Variate	Iterative weights from the generalized linear model fitting
Linear predictor	Variate	Linear predictor from the generalized linear model fitting
Adjusted response variate	Variate	Adjusted response variate
Adjusted dependent variate	Variate	Adjusted dependent variate on the linear predictor scale
Residuals on linear predictor scale	Variate	Residuals from the fit on linear predictor scale
SE of linear predictor residuals	Variate	Standard errors of linear predictor residuals
Exit status	Scalar	The exit status from the GLMM procedure. A value of 0 indicates success.
Means	Table	Predicted means for the specified model terms
Back transformed means	Table	Back transformed means for the specified model terms
SEDS of means	Symmetric matrix	Standard errors of differences between means for the specified model term
Variance-covariance matrix for means	Symmetric matrix	Variance-covariance matrix for the means for the specified model term
F statistic	Scalar	The F statistic for the specified model term using the Calculation method
Numerator degrees of freedom	Scalar	The numerator degrees of freedom for the specified model term
Denominator degrees of freedom	Scalar	The denominator degrees of freedom for the specified model term

Terms for residuals

The list allows selection from type of residuals that can be used.

All	Use the residuals combined from all random terms.
Final	Use the residuals from the final random term.

Wald statistics type

This option selects the type of Wald test used.

Add	the tests are for adding terms sequentially to the model
Drop	the tests are for dropping terms from the full fixed model

If Drop is selected, tests will only be made for terms which are not marginal to other terms in the model, e.g. for A + B + A.B, only the A.B term will be tested are A and B are marginal to A.B, and so cannot be dropped from the full model.

Terms for means and back transformed means

Specifies the model term for which tables of estimated means are to be saved. If tables of means are required for more than one model term, this menu should be invoked once for each term, changing the specification of the model term each time.

Calculation method for F-statistics

This controls whether Wald tests for fixed effects are accompanied with approximate F statistics and corresponding numbers of residual degrees of freedom. The computations, using the method devised by Kenward & Roger (1997), can be time consuming with large or complicated models. So, the default setting automatic, can be used to allow Genstat to assess the model itself and decide automatically whether to do the computations and which method to use. The other settings allow you to control what to do yourself:

none	No F statistics are produced
algebraic	F statistics are calculated using algebraic derivatives (which may involve large matrix calculations)
numerical	F statistics are calculated using numerical derivatives (which require an extra evaluation of the mixed model equations for every variance parameter).

Offset

The estimated means and back-transformed means will be estimated at the given offset. If this is not provided, the mean offset in the model will be used.

Display in spreadsheet

Select this to display the results in a new spreadsheet window. The format of the table of means and Back transformed means is controlled by the Format drop down list.

Page format	Each table is displayed on a single page in a separate spreadsheet. The last specified classifying factor indexes the columns in the spreadsheet.
Column format	The tables occupy a single column but multiple tables can be put in a single spreadsheet. This is the default for a table with a single classifying factor.

Action Icons

	Clear	Clear all fields and list boxes.
	Help	Open the Help topic for this dialog.