csummary.dmm {dmm}

Make summary tables of (co)variance component estimates sorted into class-specific classes for a dmm object.

Description

Extracts the (co)variance component estimates from an object of class dmm, for the specified set of traits and set of components. Makes tables of component estimates ordered either by trait or by component. Tables include component estimate, its standard error, and its 95 percent confidence limits. Components are grouped into class-specific classes, if there are any class-specific components fitted.

Usage

## S3 method for class 'dmm'
csummary(object, traitset = "all", componentset = "all", bytrait = T,
            gls = F, digits = 3, ...)

Arguments

object

An object of class dmm. (Co)variance component estimates are obtained from this object.

traitset

A vector containing the names of the subset of traits for which tables of (co)variance component estimates are to be constructed. Default is "all" which means all traits present in object object.

componentset

A vector containing the names of the subset of (co)variance components for which tables are to be constructed. Default is "all" which means all (co)variance components present in object object.

bytrait

Logical flag: should the tables of (co)variance component estimates be constructed with trait varying least rapidly from line to line? If TRUE each subtable contains component estimates for one trait or traitpair and for all components in argument componentset. If FALSE each subtable contains component estimates for one component and for all traits or traitpairs.

gls

Logical flag: should the (co)variance component estimates by GLS-b method be tabled in addition to the (co)variance component estimates by OLS-b method? Default is gls=FALSE. The GLS-b (co)variance component estimates can only be tabled if object object contains the attribute gls, that is if object was constructed by a dmm() call with argument gls=TRUE.

digits

Number of digits for output. This is returned as part of the return value for us e by the S3 print function print.csummary.dmm().

...

Ellipsis argument.

Details

This is a long printout with estimates, standard errors and confidence limits, arranged in tables with one estimate per line. For a short printout see function print.dmm(). In the case of class-specific components, the components are listed in class groups, so that they sum to the class phenotypic variance within each group. In the case where all components are non-specific, there is just one class group.

Value

An object of class csummary.dmm which is a list containing the following items:

ctables

A list of dataframe objects each containing one subtable of estimates of the (co)variance components, along with the appropriate standard errors and confidence limits. Based on OLS-b component estimates.

gctables

A list of dataframe objects each containing one subtable of estimates of the (co)variance components, along with the appropriate standard errors and confidence limits. Based on GLS-b component estimates. Only present if argument gls=TRUE.

traits

A vector of traitnames as specified in argument traitset.

components

A vector of component names as specified in argument componentset.

bytrait

Logical flag: as specified in argument bytrait.

gls

Logical flag: as specified in argument gls.

digits

A numeric value, as specified in argument digits.

call

The function call

Note

There is no provision to constrain the 95 percent confidence limits for component estimates. Hence for small samples, these may vary outside the bounds for the component, that is for components which are variances, they may be negative. Fixed effects are not bounded. Use csummary() if you want to see the components summing to phenotypic (co)variance, and sorted into class-specific groups. Use summary() if you just want all the components as estimated.

Author(s)

Neville Jackson

See Also

Function print.csummary.dmm().

Examples

# get some data
data(sheep.df)
# prepare it - only need "E" and "A" relationship matrices
sheep.mdf <- mdf(sheep.df,pedcols=c(1:3),factorcols=c(4:6),ycols=c(7:9),
             sexcode=c("M","F"),relmat=c("E","A"))
# estimate (co)variance components  - VarG(Ia) is Sex-specific
sheep.fitc <- dmm(sheep.mdf, Ymat ~ 1 + Year + Sex,
              components=c("VarE(I)"),
              specific.components=list(Sex=c("VarG(Ia)")))
# look  at components within Sex classes 
csummary(sheep.fitc,bytrait=FALSE)
# look just at trait "Cww"
summary(sheep.fitc,traitset="Cww")
# cleanup
rm(sheep.df)
rm(sheep.mdf)
rm(sheep.fitc)