VRKEEP procedure

Saves output for a REML fixed model fitted in a Genstat regression (R.W. Payne).

Options

Parameters

Description

VRKEEP saves output from models fitted by procedures VRFIT, VRADD, VRDROP, VRSWITCH and VRTRY, which are designed to improve the process of determining the appropriate fixed terms to include in a REML analysis. They do this by a generalized regression analysis, with a weight matrix based on variances estimated from the original REML analysis (with the full fixed model). See VRFIT for more details.

The TERMS parameter specifies terms about which you wish to save information. As in FIT, the FACTORIAL option sets a limit on the number of factors and variates in each term. Any term containing more than that limit is deleted. The subsequent parameters allow you to specify identifiers of data structures to store the various types of information for each of the terms that you have specified. The ESTIMATES parameter saves estimates for each term, in a table if the term involves factors or in a scalar if it involves only variates. Similarly the SE parameter saves standard errors for the estimates. The VCOVARIANCE parameter saves the variances and covariances between the estimates of each term, in a symmetric matrix if the term involves factors or in a scalar if it involves only variates. The NDF and DDF parameters saves the number of numerator and denominator degrees of freedom for the terms, in scalars. If you have a single term, you can supply a table, scalar or symmetric matrix for each of these parameters, as appropriate. However, if you have several terms, you must supply a pointer which will then be set up to contain as many tables, scalars or symmetric matrices as there are terms.

The RESIDUALS and FITTEDVALUES options save the residuals and fitted values, respectively. The RMETHOD option controls the type of residuals that are formed.

The RDF and RSS options save the number of residual degrees of freedom and the residual sum of squares.

The ACCUMULATED parameter saves the accumulated analysis-of-variance table, as a pointer with elements labelled ‘Change‘, ‘s.s.‘, ‘n.d.f.’, ‘m.s.’, ‘Wald‘,’F‘, ‘Wald pr.’, ‘d.d.f.’ and ‘F pr.‘. The last two elements will contain missing values if the denominator degrees of freedom of the terms could not be estimated in the original REML analysis. The pointer is defined so that the case of the labels is not significant.

The DENOMINATOR option specifies how the residual is selected for the accumulated analysis of variance By default it is taken from the model with the smallest number of residual degrees of freedom. However, you can set DENOMINATOR=ms to take it from the model with the smallest residual mean square.

Options: FACTORIAL, RESIDUALS, FITTEDVALUES, RMETHOD, RDF, RSS, ACCUMULATED, DENOMINATOR.
Parameters: TERMS, ESTIMATES, SE, VCOVARIANCE, NDF, DDF.

See also

Directives: RKEEP, REML.
Procedures: VRADD, VRDISPLAY, VRDROP, VRFIT, VRSETUP, VRSWITCH, VRTRY.
Commands for: REML analysis of linear mixed models.

Example

CAPTION     'VRKEEP example',\
            'Example 5.3.6 from The Guide to Genstat, Part 2 Statistics';\
            STYLE=meta,plain
FACTOR      [NVALUES=322; LEVELS=27] Dam
&           [NVALUES=322; LEVELS=18] Pup
FACTOR      [NVALUES=322; LEVELS=2; LABELS=!T('M','F')] Sex
FACTOR      [NVALUES=322; LEVELS=3; LABELS=!T('C','Low','High')] Dose
VARIATE     [NVALUES=322] Littersize,Weight
OPEN        '%GENDIR%/Examples/GuidePart2/Rats.dat'; CHANNEL=chan
READ        [CHANNEL=chan] Dose,Sex,Littersize,Dam,Pup,Weight; \
            FREPRESENTATION=2(labels),4(levels)
CLOSE       chan
VCOMPONENTS [FIXED=Littersize+Dose*Sex] RANDOM=Dam/Pup
REML        [PRINT=model,components,wald] Weight
VRFIT       [PRINT=estimates,accumulated] Littersize+Sex*Dose
VRKEEP      [ACCUMULATED=accumulated] Littersize+Sex+Dose;\
            ESTIMATES=!p(lest,sest,dest); SE=!p(lse,sse,dse);\
            DDF=!p(lddf,sddf,dddf)
PRINT       accumulated[]
&           lest,lse,lddf
&           sest,sse & sddf
&           dest,dse & dddf