lme4breeding (version 1.1.0)

predict.lmeb: Predict form of a LMM fitted with lmeb/lmebreed

Description

predict method for class "lmeb".

Usage

# S3 method for lmeb
predict(object, hyperTable=NULL, classify, usePEV=FALSE, ...)

Value

pvals

the table of predictions according to the specified arguments.

mapCondVar

the map between the hypertable and the effects.

hyperTable

the table specifying the terms to include and terms to be averaged.

b

the fixed and random effects in a vector form.

condVarMat

the variance covariance for the predictions.

D

the model matrix for predictions as defined in Welham et al.(2004).

classify

the character value used to indicate which term we are computing the predictions for.

Arguments

object

a mixed model of class "lmeb" fitted with lmeb

hyperTable

a data frame with columns; variable, group, type, include, average. See the Dtable function to understand the format.

A pure include term means that the model matrices for that fixed or random effect is filled with 1s for the positions where column names and row names match.

An include and average term means that the model matrices for that fixed or random effect is filled with 1/#1s in that row.

A pure average term alone means that all rows for such fixed or random effect will be filled with 1/#levels in the effect.

If a term is not considered include or average is then totally ignored in the BLUP and SE calculation.

The default behavior when the user doesn't provide the hyperTable is to include and average any fixed effect that is not part of classify. Include any term making a perfect match with the classify argument and include and average any imperfect match with classify argument (e.g., interactions).

classify

is a character value indicating which term we are computing the predictions for.

usePEV

is a logical value indicating whether we should use the conditional variance or the PEV for the computation of standard errors for the linear combination.

...

Further arguments to be passed.

Author

Giovanny Covarrubias-Pazaran

Details

This function allows to produce predictions specifying those variables that define the margins of the hypertable to be predicted (argument classify). Predictions are obtained for each combination of values of the specified variables that is present in the data set used to fit the model. See vignettes for more details.

For predicted values the pertinent design matrices X and Z together with BLUEs (b) and BLUPs (u) are multiplied and added together.

predicted.value = Db

standard.error = D Cinv Dt

References

Welham, S., Cullis, B., Gogel, B., Gilmour, A., and Thompson, R. (2004). Prediction in linear mixed models. Australian and New Zealand Journal of Statistics, 46, 325 - 347.

See Also

predict, lmeb

Examples

data(DT_yatesoats, package="enhancer")
DT <- DT_yatesoats
m3 <- lmeb(Y ~ V + N + V:N +
                 (1|B) + (1|B:MP),
               data = DT)

#############################
## predict means for nitrogen
#############################
pp=predict(object=m3, classify="N")
pp$pvals
pp$hyperTable

#############################
## predict means for variety
#############################
pp=predict(object=m3, classify="V")
pp$pvals
pp$hyperTable
#############################
## predict means for nitrogen:variety
#############################
pp=predict(object=m3, classify="N:V")
pp$pvals
ht <- pp$hyperTable
ht[4,"include"]=1
ht[4,"average"]=0

pp2=predict(object=m3, classify="N:V", hyperTable=ht)
pp2$pvals