JackLmer: Jackknife components for normal data

In fullfact: Full Factorial Breeding Analysis

Jackknife components for normal data

Description

Extracts additive genetic, non-additive genetic, and maternal variance components from a linear mixed-effect model using the lmer function of the lme4 package. Model random effects are dam, sire, and dam by sire.

Usage

JackLmer(observ, dam, sire, response, ml = F, size = 1, first = NULL)

Arguments

observ	Data frame of observed data.
dam	Column name containing dam (female) parent identity information.
sire	Column name containing sire (male) parent identity information.
response	Column name containing the offspring (response) phenotype values.
ml	Default is FALSE for restricted maximum likelihood. Change to TRUE for maximum likelihood.
size	Default is 1 for delete-one jackknife resampling. If size > 1, delete-d jackknife resampling occurs removing a block d equal to size.
first	Number of initial sub-samples to run. Useful for examing if there is variation among sub-samples before jackknife resampling the entire data set. There can be little variation for delete-one jackknife resampling with large data sets, and delete-d jackknife resampling should be considered.

Details

Uses delete-one jackknife resampling (Efron & Tibshirani 1993, p. 141-145). For the option of delete-d jackknife resampling, the rows of the observed data frame are shuffled and a block of observations of size d is deleted sequentially. Extracts the dam, sire, dam, dam by sire, and residual variance components. Calculates the total variance component. Calculates the additive genetic, non-additive genetic, and maternal variance components (see Lynch and Walsh 1998, p. 603).

Value

A data frame with columns containing the raw variance components for dam, sire, dam by sire, residual, total, additive genetic, non-additive genetic, and maternal. The number of rows in the data frame matches the total number of observations (N) for delete-one jackknife resampling or M groups for delete-d jackknife resampling to the lowest integer. Each row represents a deleted single observation or deleted d observations group.

Note

Maximum likelihood (ML) estimates the parameters that maximize the likelihood of the observed data and has the advantage of using all the data and accounting for non-independence (Lynch and Walsh 1998, p. 779; Bolker et al. 2009). On the other hand, ML has the disadvantage of assuming that all fixed effects are known without error, producing a downward bias in the estimation of the residual variance component. This bias can be large if there are lots of fixed effects, especially if sample sizes are small. Restricted maximum likelihood (REML) has the advantage of not assuming the fixed effects are known and averages over the uncertainty, so there can be less bias in the estimation of the residual variance component. However, REML only maximizes a portion of the likelihood to estimate the effect parameters, but is the preferred method for analyzing large data sets with complex structure.

References

Efron B, Tibshirani R. 1993. An introduction to the Bootstrap. Chapman and Hall, New York.

Lynch M, Walsh B. 1998. Genetics and Analysis of Quantitative Traits. Sinauer Associates, Massachusetts.

See Also

JackLmer2, JackLmer3

Examples

data(chinook_length) #Chinook salmon offspring length

#Delete-one
#length_jack1<- JackLmer(observ=chinook_length,dam="dam",sire="sire",response="length")
length_jack1<- JackLmer(observ=chinook_length,dam="dam",sire="sire",response="length",
first=2) #first 2

#Delete-d, d=5
#length_jackD<- JackLmer(observ=chinook_length,dam="dam",sire="sire",response="length",
#size=5)
length_jackD<- JackLmer(observ=chinook_length,dam="dam",sire="sire",response="length",
size=5,first=2) #first 2

fullfact documentation built on May 29, 2024, 1:21 a.m.