In charlotte-ngs/asmss2022: Applied Statistical Methods in Animal Science - Spring Semester 2022

Disclaimer

Data on slaughter weight and sex used in Problem 1 are generated.

Assumptions

Constants and effects are specified

n_mean_sw <- 287.3
n_sd_sw <- 9.5
n_eff_castrate <- 7.1
n_eff_female <- -13.8
n_eff_male <- 5.9
# true vector of effects
vec_eff_sex <- c(n_mean_sw, n_eff_castrate, n_eff_female, n_eff_male)
# tibble without data, but with assignment of sex to animals
n_nr_rec <- 12
tbl_sw <- tibble::tibble(Animal = c(1:n_nr_rec),
                         Sex    = c(rep("female", 4),
                                    rep("castrate", 4),
                                    rep("male", 4)),
                         `Slaughter Weight` = rep(0, n_nr_rec))
# model matrix
mat_X <- model.matrix(lm(`Slaughter Weight` ~ 0 + Sex, data = tbl_sw))
attr(mat_X, "assign") <- NULL
attr(mat_X, "contrasts") <- NULL
colnames(mat_X) <- NULL
mat_X <- cbind(matrix(1, nrow = n_nr_rec, ncol = 1), mat_X)
mat_X

Use the design matrix, the effects and the resiudals to produce the data vector.

set.seed(6723)
mat_sw <- mat_X %*% vec_eff_sex + rnorm(n_nr_rec, mean = 0, sd = n_sd_sw)
mat_sw

Add observations to tibble

tbl_sw$`Slaughter Weight` <- round(mat_sw[,1], digits = 1)
tbl_sw

Write the data to a csv-file

s_exam_data_p01 <- file.path(here::here(), "docs", "data", "asm_exam_p01.csv")
if (!file.exists(s_exam_data_p01))
  readr::write_csv(tbl_sw, s_exam_data_p01)

Fitting the Linear Model

lm_sw <- lm(`Slaughter Weight` ~ Sex, data = tbl_sw)
summary(lm_sw)

charlotte-ngs/asmss2022 documentation built on June 7, 2022, 1:33 p.m.