Saves results from a REML analysis in a spreadsheet (R.W. Payne).
Options
COMPONENTS = variate |
Variate to contain the variance components; default components |
|---|---|
MEANS = pointer |
Pointer to tables to contain the means; default means |
SEDMEANS = pointer |
Pointer to matrices to contain the standard errors of differences of the means; default sedmeans |
VARMEANS = pointer |
Pointer to matrices to contain the variance-covariance matrices of the means; default varmeans |
EFFECTS = pointer |
Pointer to tables to contain the effects; default effects |
SEDEFFECTS = pointer |
Pointer to matrices to contain the standard errors of differences of the effects; default sedeffects |
VAREFFECTS = pointer |
Pointer to matrices to contain the variance-covariance matrices of the effects; default vareffects |
REPLICATIONS = pointer |
Pointer to tables of replications; default replication |
WALDTABLE = pointer |
Pointer to a text and variates containing the information in the table of tests for fixed effects; default waldtable |
PTERMS = formula |
Terms (fixed or random) for which effects or means are to be saved; default * implies all the fixed terms |
FMETHOD = string token |
Controls whether and how to calculate F-statistics for fixed terms (automatic, none, algebraic, numerical); default auto |
SPREADSHEET = string tokens |
What to include in the spreadsheet (components, waldtable, effects, sedeffects, vareffects, means, sedmeans, varmeans, replications); default comp, wald, mean, sedm, repl |
OUTFILENAME = text |
Name of Genstat workbook file (.gwb) or Excel (.xls or .xlsx) file to create |
SAVE = REML save structure |
Specifies which REML analysis to save; default * i.e. most recent one |
No parameters
Description
VSPREADSHEET puts results from a REML analysis into a spreadsheet. By default the results are from the most recent REML, but you use the SAVE option to specify the save structure from some other analysis.
The SPREADSHEET option specifies which pages of the spreadsheet to form, with settings:
components |
variance components, |
|---|---|
waldtable |
tests for fixed effects, |
effects |
tables of effects, |
sedeffects |
standard errors of differences of effects, |
vareffects |
variance-covariance matrices of effects, |
means |
tables of means, |
sedmeans |
standard errors of differences of means, |
varmeans |
variance-covariance matrices of means, |
replications |
replication tables. |
(Note: this includes only the information readily assembled from VKEEP. So, for example, parameters of correlation models are not available.) By default, SPREADSHEET = comp, wald, mean, sedm, repl.
To help avoid clashes between the columns of the spreadsheets if you want to save results from more than one analysis, the parameters COMPONENTS, WALDTABLE, EFFECTS, SEDEFFECTS, VAREFFECTS, MEANS, SEDMEANS, VARMEANS and REPLICATIONS allow you to specify identifiers for the columns (or sets of columns) that will store the corresponding results in the current spreadsheet.
You can save the data 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, the data are put into a spreadsheet opened inside Genstat.
Options: COMPONENTS, MEANS, SEDMEANS, VARMEANS, EFFECTS, SEDEFFECTS, VAREFFECTS, REPLICATIONS, WALDTABLE, PTERMS, FMETHOD, SPREADSHEET, OUTFILENAME, SAVE.
Parameters: none.
Action with RESTRICT
If the Y variate is restricted, that restriction will carry over into the fitted-values spreadsheet.
See also
Directive: SPLOAD.
Procedures: ASPREADSHEET, AUSPREADSHEET, RSPREADSHEET, FSPREADSHEET.
Commands for: REML analysis of linear mixed models.
Example
CAPTION 'VSPREADSHEET example',\
'Incomplete block analysis (from Cochran & Cox 1957, p.406).';\
STYLE=meta,plain
FILEREAD [NAME='%gendir%/examples/REML-1.DAT']\
Reps,Blocks,Treats,Yield; FGROUPS=3(yes),no
FACTOR [LEVELS=5; VALUES=(1...5)10] Plots
VCOMPONENTS [FIXED=Treats] RANDOM=Reps/Blocks/Plots
REML [PRINT=components,means,wald; PSE=alldiff] Yield
VSPREADSHEET