Forms a model-definition structure for a REML analysis (R.W. Payne).
Options
MODELSTRUCTURE = pointer |
Specifies the model-definition structure; no default (must be specified) |
|---|---|
DESCRIPTION = text |
Description of the model (for output) |
FIXED = formula |
Fixed model terms; default * |
CONSTANT = string token |
How to treat the constant term (estimate, omit); default esti |
FACTORIAL = scalar |
Limit on the number of factors or covariates in each fixed term; default 3 |
CADJUST = string token |
What adjustment to make to covariates before analysis (mean, none); default mean |
CHANGEITEMS = string tokens |
What changes to make to an existing model-definition structure (description, fixed, constant, factorial, cadjust, random, initial, constraints); if this is unset, the structure is redefined completely |
IMODELSTRUCTURE = pointer |
Specifies the initial model-definition structure, to modify when CHANGEITEMS is set; default is to modify the one specified by MODELSTRUCTURE |
EXPERIMENTS = factor |
Factor defining the different experiments in a multi-experiment (meta-) analysis |
Parameters
RANDOM = formula |
Random model terms |
|---|---|
INITIAL = scalars |
Initial values for each component |
CONSTRAINTS = string tokens |
How to constrain each variance component and the residual variance (none, positive, fixrelative, fixabsolute); must be set unless MODIFY=yes |
Description
VFMODEL is one of a suite of procedures designed to simplify the assessment of alternative models for a REML analysis. The first step is to form a model-definition structure for each candidate model, using the VFMODEL and VFSTRUCTURE procedures (these define the model settings controlled by the VCOMPONENTS and VSTRUCTURE directives, respectively). The model-definition structures can then be used as input to procedures like VARANDOM, which assesses possible random models. VARANDOM uses VMODEL to specify each model, in turn, so that it can fit it using REML. The relevant results from each fit are saved by the VRACCUMULATE procedure, so that a decision about the recommended random model can be made once they have all been tried.
The model-definition structure is specified by the MODELDEFINITION option, which must be set. The DESCRIPTION option supplies a (brief, one-line) description to identify the model in the output.
Details of the model are specified by the FIXED, CONSTANT, FACTORIAL, CADJUST and EXPERIMENTS options, and the RANDOM, INITIAL, CONSTRAINTS parameters (which correspond to those options and parameters of the VCOMPONENTS directive).
You can set the CHANGEITEMS option to modify an existing model-definition structure, instead of redefining it. Its settings then specify which aspects are to be changed. By default, the existing definition structure is supplied by MODELSTRUCTURE (and the modified structure replaces the existing one). Alternatively, if you want to keep the existing structure, you can specify it separately, using the IMODELSTRUCTURE option.
Options: MODELSTRUCTURE, DESCRIPTION, FIXED, CONSTANT, FACTORIAL, CADJUST, CHANGEITEMS, IMODELSTRUCTURE, EXPERIMENTS.
Parameters: RANDOM, INITIAL, CONSTRAINTS.
See also
Directives: REML, VCOMPONENTS, VSTRUCTURE.
Procedures: VARANDOM, VFSTRUCTURE, VMODEL.
Commands for: REML analysis of linear mixed models.
Example
CAPTION 'VFMODEL example',\
'Slate Hall Farm data (Guide to REML in Genstat, Section 1.8).';\
STYLE=meta,plain
SPLOAD '%gendir%/data/slatehall.gsh'
" define model for analysis as a Lattice square design "
VFMODEL [MODELSTRUCTURE=Latticesq; DESCRIPTION='Lattice square analysis';\
FIXED=variety] replicates/(rows*columns)
VMODEL Latticesq
REML yield