Finds the best REML random model from a set of models defined by VFMODEL (R.W. Payne).
Options
PRINT = string tokens |
Controls what summary output is produced about the models (deviance, aic, bic, sic, dffixed, dfrandom, change, exit, best); default devi, aic, sic, dfra, best |
|---|---|
PBEST = string tokens |
Controls the output from the REML analysis with the best model (model, components, effects, means, stratumvariances, monitoring, vcovariance, deviance, Waldtests, missingvalues, covariancemodels, aic, sic, bic); default * i.e. none |
PTRY = string tokens |
Controls the output to present to present from the REML analysis used to try each model (model, components, effects, means, stratumvariances, monitoring, vcovariance, deviance, Waldtests, missingvalues, covariancemodels, aic, sic, bic); default * i.e. none |
MODELSTRUCTURES = pointer |
Model-definition structures specifying the models to try |
PTERMS = formula |
Terms (fixed or random) for which effects or means are to be printed; default * implies all the fixed terms |
PSE = string token |
Standard errors to be printed with tables of effects and means (differences, estimates, alldifferences, allestimates, none); default diff |
MVINCLUDE = string tokens |
Whether to include units with missing values in the explanatory factors and variates and/or the y-variates (explanatory, yvariate); default * i.e. omit units with missing values in either explanatory factors or variates or y-variates |
METHOD = string token |
How to choose the best model (aic, sic, bic); default sic |
PROHIBIT = string token |
Whether to exclude models where any estimated variance parameters are held at a bound (bound); default * |
Parameters
Y = variates |
Response variates |
|---|---|
NBESTMODEL = scalars |
Saves the number of the best model for each y-variate, returning a missing value if no models could be fitted successfully |
SAVE = REML save structures |
Save structure from the analysis of the best model for each y-variate |
Description
VARANDOM allows you to try several alternative random models for a REML analysis, and then select the best one according to either their Akaike or Schwarz (Bayesian) information coefficients.
Model-definition structures for the models to be assessed must be specified using the MODELSTRUCTURES option. These are formed using the VFMODEL and VFSTRUCTURE procedures, which define the aspects controlled by the VCOMPONENTS and VSTRUCTURE directives, respectively.
The response variate for the analysis must be specified by the Y parameter. The number of the best model can be saved, in a scalar, by the NBESTMODEL parameter; it returns a missing value if no models could be fitted successfully The REML save structure from the analysis with the best model can be saved using the SAVE parameter.
The MVINCLUDE option controls whether units with missing values in the explanatory factors and variates and/or the y-variate are included in the analysis, as in the REML directive.
The METHOD option specifies how to assess the models
aic |
uses their Akaike information coefficients, |
|---|---|
sic or bic |
uses their Schwarz (Bayesian) information coefficients (default). |
You can set option PROHIBIT = bound, to excludes models with any estimated variance parameters held at a bound.
The PRINT option specifies the summary output to be produced about the models. The settings are mainly the same as those of the VRACCUMULATE procedure (which is used to store and then print details of the analyses). There is also an extra setting best, which prints the description of the best model. The default is to print the best description, together with the deviance, the Akaike and Schwarz (Bayesian) information coefficients and the number of degrees, for all the random models.
The PBEST option specifies the output to be produced from the REML analysis with the best model. Similarly, the PTRY option indicates what output should be produced for each candidate random model when it is tried. Their settings are mainly the same as those of the PRINT option of the REML directive. There are also extra settings aic and sic (with a synonym bic) to print the Akaike and Schwarz (Bayesian) information coefficients, respectively. The default for both these options is to produce no output.
The PTERMS option operates as in REML, to specify the terms whose means and effects are printed by PBEST and PTRY; the default is all the fixed terms. Likewise, the PSE option controls the type of standard error that is displayed with the means and effects; the default is to give a summary of the standard errors of differences.
Options: PRINT, PBEST, PTRY, MODELSTRUCTURES, PTERMS, PSE, MVINCLUDE, METHOD, PROHIBIT.
Parameters: Y, NBESTMODEL, SAVE.
See also
Directives: REML, VCOMPONENTS, VSTRUCTURE.
Procedures: VAOPTIONS, VARECOVER, VFMODEL, VFSTRUCTURE, VMODEL, VRACCUMULATE.
Commands for: REML analysis of linear mixed models.
Example
CAPTION 'VARANDOM 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 randomized-blocks design "
VFMODEL [MODELSTRUCTURE=RCBD; DESCRIPTION='Randomized blocks';\
FIXED=variety] replicates
" define model for analysis as a Lattice square design "
VFMODEL [MODELSTRUCTURE=Latticesq; DESCRIPTION='Lattice square';\
FIXED=variety] replicates/(rows*columns)
" define model for analysis with an AR1 (x) AR1 model "
VFMODEL [MODELSTRUCTURE=AR1xAR1; DESCRIPTION='AR1 (x) AR1';\
FIXED=variety] fieldrow.fieldcolumn
VFSTRUCTURE [MODELSTRUCTURE=AR1xAR1; TERMS=fieldrow.fieldcolumn]\
2('AR'); ORDER=1; FACTOR=fieldrow,fieldcolumn
" define model for analysis with an AR1 (x) AR1 model + measurement error "
VFMODEL [MODELSTRUCTURE=AR1xAR1p; DESCRIPTION='AR1 (x) AR1 + plots';\
FIXED=variety] fieldrow.fieldcolumn+plotnumber
VFSTRUCTURE [MODELSTRUCTURE=AR1xAR1p; TERMS=fieldrow.fieldcolumn]\
2('AR'); ORDER=1; FACTOR=fieldrow,fieldcolumn
VARANDOM [MODELSTRUCTURES=RCBD,Latticesq,AR1xAR1,AR1xAR1p]\
yield; SAVE=savebest
VDISPLAY [PRINT=model,components,wald] savebest