aiCI {nadiv} | R Documentation |
Confidence Intervals for Variance Components
Description
Produces the 1-alpha Upper and Lower Confidence Limits for the variance components in an ASReml-R model.
Usage
aiCI(asr.model, Dimnames = NULL, alpha = 0.05)
Arguments
asr.model |
Object from a call to |
Dimnames |
A vector of characters if names are desired for the output.
If not specified, the default labels from the |
alpha |
A numeric value indicating the level of Type I error for constructing the Confidence Intervals. |
Details
Variances from the inverse of the Average Information matrix of an ASReml
model are translated according to the varTrans
function and
used in constructing the 1-alpha Confidence Interval.
Value
A matrix
is returned with a row for each variance component.
The three columns correspond to the Lower Confidence Limit, estimate from
the asreml
model, and Upper Confidence Limit for each variance
component.
Note
The vector of Dimnames
should match the same order of variance
components specified in the model.
Author(s)
See Also
Examples
## Not run:
library(asreml)
ginvA <- ainverse(warcolak)
ginvD <- makeD(warcolak[, 1:3])$listDinv
attr(ginvD, "rowNames") <- as.character(warcolak[, 1])
attr(ginvD, "INVERSE") <- TRUE
warcolak$IDD <- warcolak$ID
warcolak.mod <- asreml(trait1 ~ sex,
random = ~ vm(ID, ginvA) + vm(IDD, ginvD),
data = warcolak)
summary(warcolak.mod)$varcomp
aiCI(warcolak.mod)
## End(Not run)