WOMBAT Example 19

Simulated data set for 9 traits to illustrate penalized estimation for
`simple' penalties derived assuming a beta distribution.

Strength of penalization is regulated by a single parameter, ESS.
Suggested values for a mild penalty are 4 to.

Paternal half-sib design for 400 sires with 10 progeny each; simple
animal model with means as only fixed effect.

0: Parameter file to simulate data

A: Standard *un*penalized analysis (for comparison)

B: *Un*penalized analysis, parameterised to elements of the canonical
   decomposition. This run shows that this parameterisation to 
   canonical eigenvalues tends to have a slower convergence rate 
   than the standard form.

C: Penalty on canonical eigenvalues with ESS=8

D: Penalty on genetic partial correlations with ESS=8, shrinking
   partial genetic correlation towards their phenotypic counterparts

Z: Script to simulate replicates & Fortran code to summarize results;
   these allow simulations in Genetics 203:1885-1900 to be replicated.

