AYPARALLEL procedure

Does the same analysis of variance for several y-variates, and collates the output (R.W. Payne & D.B. Baird).

Options

Parameters

Description

The AYPARALLEL procedure does a “parallel” analysis of variance for several y-variates, combining and summarizing the information from all the analyses. The procedure operates most efficiently if the y-values for the analyses are in separate variates (with their units in identical orders). The variates should be placed into a pointer, which should then be used as the setting of the Y parameter of AYPARALLEL. The alternative format has the values for all the analyses in a single variate (which should again be used as the setting of the Y parameter). You must then also set the VFACTOR parameter, to a factor to indicate which values are involved in each analysis.

The BLOCKSTRUCTURE and TREATMENTSTRUCTURE options can specify block and treatment formulae (as in ordinary ANOVA) to define the models for the analysis of variance. If the TREATMENTSTRUCTURE option is not set, AYPARALLEL will use the model already defined by the TREATMENTSTRUCTURE directive, or will fail if that too has not been set. Similarly, if the BLOCKSTRUCTURE option is not set, AYPARALLEL will use the model (if any) previously defined by the BLOCKSTRUCTURE directive. The lengths of the block and treatment factors should be the same as the Y variate or variates. Furthermore, if there is a single Y variate, there must be a unique combimation of the levels of the block factors for every unit (this is required so that AYPARALLEL can check that the y-values are in the correct order for each analysis). The FACTORIAL option sets a limit on the number of factors in a treatment term, as in the ANOVA directive. Similarly the CONTRASTSLIMIT and DEVIATIONSLIMIT options operate as the CONTRASTS and DEVIATIONS options of ANOVA.

The COVARIATE option can list any covariates for the analyses; if this is unset, the default is taken from any existing setting defined by the COVARIATE directive. The lengths of the covariates should be the same as the Y variate or variates.

The RESIDUALS and FITTEDVALUES parameters can save the residuals and fitted values, respectively. These will each be in a pointer with a variate for each analysis if the y-values were specified in separate variates, or in a single variate if the y-values were combined in a single variate. The REPLICATION option saves a pointer containing the replication tables for the SAVETERMS. Parameters MEANS and EFFECTS save tables of means and effects from each analysis. The information is stored in a pointer with a matrix for each of the SAVETERMS. The matrices have a row for each analysis, and the columns are labelled to show how they correspond to the cells of the table. (Note that their ordering is the same as the order in which the contents of the REPLICATION table is stored.) Similarly SEEFFECTS saves effective standard errors for the effects, and VCMEANS saves the variances and covariances of the means. VAREFFECTS saves a pointer of variates storing the unit variances of the effects, obtained by the VARIANCE parameter of AKEEP. Parameters DF, SS, MS, RDF, RSS, RMS, VR and PRVR store information from the analysis of variance table, in pointers with a variate for each term and a unit for each analysis. DF store the number of degrees of freedom for the relevant term (and analysis), SS stores sums of squares, MS stores mean squares, VR stores variance ratios, and PRVR the corresponding probabilities. Similarly the RDF parameter stores the number of degrees of freedom for the appropriate residual for the term, RSS stores the residual sums of squares, and RMS the residual mean square.

Printed output is controlled by the PRINT option, with settings:

The SPREADSHEET option lets you save various output components in spreadsheets. You can save these in either a Genstat workbook (.gwb) or an Excel spreadsheet (.xls or .xlsx), by setting the OUTFILENAME option to the name of the file to create. If the name is specified without a suffix, ‘.gwb’ is added (so that a Genstat workbook is saved). If OUTFILENAME is not specified, they are put into a spreadsheet opened inside Genstat.

Options: PRINT, TREATMENTSTRUCTURE, BLOCKSTRUCTURE, COVARIATE, FACTORIAL, SAVETERMS, REPLICATION, SPREADSHEET, CONTRASTSLIMIT, DEVIATIONSLIMIT.

Parameters: Y, VFACTOR, RESIDUALS, FITTEDVALUES, MEANS, VCMEANS, EFFECTS, VAREFFECTS, SEEFFECTS, DF, SS, MS, RDF, RSS, RMS, VR, PRVR, CONTRASTS, SECONTRASTS, TCONTRASTS, PRCONTRASTS, OUTFILENAME.

Method

The analyses are performed by the ANOVA directive.

Action with RESTRICT

Any restrictions on the y-variates will be removed.

See also

Directive: ANOVA.
Procedures: MAANOVA, RYPARALLEL.
Commands for: Analysis of variance.

Example

CAPTION    'AYPARALLEL example'
VARIATE    [NVALUES=9] AB011264,AB011266,AB019525
READ       AB011264
4.2 5.1 11.6 3 3.9 9.3 3.3 4.4 10 :
READ       AB011266
12.5 3 8.8 8.7 2.3 7.1 10.3 2.5 6.9 :
READ       AB019525
1.5 6.2 8.9 1.3 4.5 7.7 1.3 5.1 7 :
FACTOR     [LEVELS=3; VALUES=(1...3)3] Reps
FACTOR     [LEVELS=!(0,1,2); ; VALUES=3(0...2)] Times
TREATMENTSTRUCTURE POL(Times;1)
BLOCKSTRUCTURE Reps
AYPARALLEL [PRINT=monitoring,summary] !p(AB011264,AB011266,AB019525);\
           RESIDUALS=residuals; FITTEDVALUES=fittedvals;\
           DF=df; SS=ss; MS=ms; RSS=rss; RDF=rdf; RMS=rms; VR=vr; PRVR=prvr;\
           EFFECTS=effects; SEEFFECTS=seeffects; TEFFECTS=teffects;\
           PREFFECTS=preffects; CONTRASTS=contrasts; SECONTRASTS=secontrasts;\
           TCONTRASTS=tcontrasts;\
           PRCONTRASTS=prcontrasts
PRINT      #df,#ss,#ms,#rdf,#rss,#rms,#vr,#prvr
&          [SERIAL=yes] residuals,fittedvals,\
           effects[],seeffects[],teffects[],preffects[],\
           contrasts[],secontrasts[],tcontrasts[],prcontrasts[]