VCA (version 1.3.1)

ranef.VCA: Extract Random Effects from 'VCA' Object.

Description

Extract random effects and possibly apply a transformation to them (standardization, studentization).

Usage

"ranef"(object, term = NULL, mode = c("raw", "student", "standard"), quiet = FALSE, ...)

Arguments

object

(VCA) object where random effects shall be extracted

term

(character) string specifying a term (factor) for which random effects should be extracted, one can also specify an integer which is interpreted as i-th element of 'obj$res.assign$terms'

mode

(character) string or abbreviation specifying whether "raw" residuals should be returned or a transformed version c("student" or "standard")

quiet

(logical) TRUE = will suppress any warning, which will be issued otherwise

...

additional parameters

Details

Extracting the 'RandomEffects' element of an 'VCA' object if this exists and applying standardization (mean 0, sd 1) or studentization. For studentized random effects the i-th random effects is divided by the i-th main diagonal element of matrix $O = GZ'QZG$, where $G$ is the covariance-matrix of random effects, $Z$ is a design matrix assigning random effects to observations and matrix $Q = V"(I - H)$ (see residuals.VCA for further details).

References

Searle, S.R, Casella, G., McCulloch, C.E. (1992), Variance Components, Wiley New York

Laird, N.M., Ware, J.H., 1982. Random effects models for longitudinal data. Biometrics 38, 963-974.

Schuetzenmeister, A. and Piepho, H.P. (2012). Residual analysis of linear mixed models using a simulation approach. Computational Statistics and Data Analysis, 56, 1405-1416

Examples

## Not run: 
# data(dataEP05A2_1)
# fit <- anovaVCA(y~day/run, dataEP05A2_1)
# ranef(fit)
# 
# # get variable-specific random effects (REs)
# # both extract the same REs
# ranef(fit, "day")
# ranef(fit, 1)
# 
# # get standardized REs
# ranef(fit, "day:run", "standard")
# 
# # or studentized REs
# ranef(fit, 2, "stu")
# ## End(Not run)