Copies information from an analysis by `VASERIES`

into Genstat data structures (R.W. Payne).

### Options

`EXPERIMENT` = scalar or text |
Specifies the experiment, from the series, whose output is to be saved; no default, must be set |
---|---|

`FACTORIAL` = scalar |
Limit on the number of factors or covariates in the terms generated from the `TERMS` parameter; default 3 |

`RESIDUALS` = variate |
Residuals from the analysis |

`FITTEDVALUES` = variate |
Fitted values from the analysis |

`DEVIANCE` = scalar |
Residual deviance from fitting the full fixed model |

`DF` = scalar |
Residual degrees of freedom after fitting the full fixed model |

`AIC` = scalar |
Saves the Akaike information coefficient |

`SIC` = scalar |
Saves the Schwarz (Bayesian) information coefficient |

`RMETHOD` = string token |
Which random terms to use when calculating `RESIDUALS` (`final` , `all` ); default `fina` |

`FMETHOD` = string token |
Controls how to calculate F-statistics for fixed terms (`automatic` , `none` , `algebraic` , `numerical` ); default `auto` |

`WMETHOD` = string token |
Controls which Wald statistics are saved (`add` , `drop` ); default `drop` |

`MODELDEFINITIONS` = pointer |
Definitions of the models used by `VASERIES` |

`SAVE` = pointer |
`REML` save structures from the `VASERIES` analysis |

### Parameters

`TERMS` = formula |
Terms for which information is to be saved |
---|---|

`COMPONENTS` = scalars |
Estimated variance components |

`MEANS` = tables |
Table of predicted means for each term |

`SEDMEANS` = symmetric matrices |
Standard errors of differences between the predicted means |

`VARMEANS` = symmetric matrices |
Variance-covariance matrix of the means |

`EFFECTS` = tables |
Table of estimated regression coefficients for each term |

`SEDEFFECTS` = symmetric matrices |
Standard errors of differences between the estimated parameters of each term |

`VAREFFECTS` = symmetric matrices |
Variance-covariance matrix of the effects of a term |

`WALD` = scalars |
Wald statistic (fixed terms only) |

`FSTATISTIC` = scalars |
F statistics (fixed terms only) |

`NDF` = scalars |
Numerator d.f. (fixed terms only) |

`DDF` = scalars |
Denominator d.f. (fixed terms only) |

### Description

`VASKEEP`

copies results from the analysis of a series of experiments, analysed by `VASERIES`

, into Genstat data structures. The model definitions and `REML`

save structures from the analysis can be specified by the `MODELDEFINITIONS`

and `SAVE`

options. If either of these is not specified, the output is taken from the most recent `VASERIES`

analysis.

The `EXPERIMENT`

option specifies the experiment from the series, whose output is to be saved. This can be set to a scalar, or to a single-valued text if the experiments factor has labels.

`VASKEEP`

caters for only the most commonly required types of output, as most of the output can be saved by `VKEEP`

in the usual way: the `SAVE`

parameter of `VASERIES`

saves a pointer of `REML`

save structures, one for each experiment in their order in the levels of the experiments factor. However, `VASKEEP`

has the advantage that it takes account of the additional complication that, to analyse each experiment, `VASERIES`

uses subset factors and variates containing only the data from that experiment. So, `VASKEEP`

automatically makes the conversion from these subset vectors to those in the full data set. The variates of residuals and fitted values will thus have the same number of units as the original y-variate, with missing values in the units belonging to the other experiments. Similarly the tables of means and effects will have values for all levels of relevant factors, with missing values for those that were absent in the current experiment. An exception, though, is that the symmetric matrices of standard errors of differences or variances will have units only for comparisons between means or effects for levels of factors that actually occurred in the experiment.

Options `RESIDUALS`

, `FITTEDVALUES`

, `DEVIANCE`

, `DF`

, `RMETHOD`

, `FMETHOD`

, `WMETHOD`

, and parameters `TERMS`

, `COMPONENTS`

, `MEANS`

, `SEDMEANS`

, `VARMEANS`

, `EFFECTS`

, `SEDEFFECTS`

, `VAREFFECTS`

, `WALD`

, `FSTATISTIC`

, `NDF`

, `DDF`

operate as in `VKEEP`

In addition. there is a `FACTORIAL`

option to set a limit on the number of factors or covariates in the terms generated from the `TERMS`

parameter (default 3). There are also options `AIC`

, `SIC`

to save the Akaike and Schwarz (Bayesian) information coefficients, respectively.

Options: `EXPERIMENT`

, `FACTORIAL`

, `RESIDUALS`

, `FITTEDVALUES`

, `DEVIANCE`

, `DF`

, `AIC`

, `SIC`

, `RMETHOD`

, `FMETHOD`

, `WMETHOD`

, `MODELDEFINITIONS`

, `SAVE`

.

Parameters: `TERMS`

, `COMPONENTS`

, `MEANS`

, `SEDMEANS`

, `VARMEANS`

, `EFFECTS`

, `SEDEFFECTS`

, `VAREFFECTS`

, `WALD`

, `FSTATISTIC`

, `NDF`

, `DDF`

.

### See also

Directives: `REML`

, `VKEEP`

.

Procedures: `VASERIES`

, `VASDISPLAY`

, `VASMEANS`

.

Commands for: REML analysis of linear mixed models.

### Example

CAPTION 'VASKEEP example'; STYLE=meta SPLOAD '%gendir%/examples/Vaseries.gsh' VASERIES [PRINT=best,deviance,aic,bic,sic,dfrandom,summary;\ PBEST=model,components,wald; FIXED=entry; EXPERIMENTS=site;\ ROWS=row; COLUMNS=column; BLOCKS=block; MVINCLUDE=yvariate;\ TRYSPATIAL=ifregular; TRYTRENDS=yes; RSTRATEGY=fastoptimal;\ VCONSTRAINTS=positive] yield; MODELDEFINITIONS=modeldefs; SAVE=save VASKEEP [EXPERIMENT='G'; AIC=aic; SIC=sic] entry;\ WALD=wald; FSTATISTIC=fstat; NDF=ndf; DDF=ddf PRINT aic,sic & wald,fstat,ndf,ddf