LBG - FS2024 – Exercise 5

Author

Peter von Rohr

Problem 1: Sire Model

As already shown during the lecture for the small data set at

https://charlotte-ngs.github.io/lbgfs2024/data/small_beef_data.csv

a sire model is to be used for the following dataset, already used in Exercise 4. The data set is available from

https://charlotte-ngs.github.io/lbgfs2024/data/beef_data_bc.csv

The ration \(\lambda_s\) between the residual variance and the sire variance can be assumed to be

\[\lambda_s = \frac{\sigma_e^2}{\sigma_s^2} = 1\]

Solution

  • Design matrix \(X\)
  • Design matrix \(Z\)
  • Coefficient matrix of mixed model equations
  • Right-hand side of mixed model equations
  • Solve mixed model equations
  • Identify estimates of fixed effects and predictions of sire effects