In charlotte-ngs/lbgfs2021: Livestock Breeding and Genomics (ETHZ) - Fall Semester 2021

knitr::opts_chunk$set(echo = TRUE)

Data

The same dataset is used as for the explanation of the Mixed Model Equation for the sire model. The data can be read from

s_data_url <- "https://charlotte-ngs.github.io/lbgfs2021/data/beef_data_bc.csv"
tbl_beef_bc <- readr::read_csv(file = s_data_url)
tbl_beef_bc$Herd <- as.factor(tbl_beef_bc$Herd)
tbl_beef_bc$Sire <- as.factor(tbl_beef_bc$Sire)
head(tbl_beef_bc)

Model

library(lme4)
fit_beef_bc_all <- lmer(formula = `Weaning Weight` ~ Herd + `Breast Circumference` + (1|Sire), data = tbl_beef_bc)
isSingular(fit_beef_bc_all)

Reducing model complexity

fit_beef_sire_no_inter <- lmer(formula = `Weaning Weight` ~  0 + (1|Sire), data = tbl_beef_bc)
isSingular(fit_beef_sire_no_inter)

Model summary

summary(fit_beef_sire_no_inter)

Data from lme4 Vignette

The package functionality is tested with a dataset that is documented to be working.

library(lme4)
str(sleepstudy)

Descriptive statistics reveals a linear dependence of Reaction on Days but each subject seams to have a different slope in the relationship. This is modelled with the following statement.

fm1 <- lmer(Reaction ~ Days + (Days | Subject), sleepstudy)
summary(fm1)

charlotte-ngs/lbgfs2021 documentation built on Dec. 19, 2021, 3:01 p.m.