Back to index.html.

Common environmental effects as random

Model

In this example, the author considers a simple maternal model in which the maternal effects (\(c\)) are not correlated with one another. The variance components include the animal genetic variance \(\sigma_a^2 = 20\), the maternal variance \(\sigma_c^2 = 15\), and the residual variance \(\sigma_e^2 = 65\).

Files

The data file (mr52.dat) contains the complete table shown in the textbook (p. 95). The columns are as follows.

1. Animal ID (piglet)
2. Sire ID
3. Dam ID
4. Sex (1 = male and 2 = female)
5. Weaning weight (kg)

The pedigree (mr52.ped) can be derived from the above data.

The parameter file should contain two random effects.

DATAFILE
mr52.dat
NUMBER_OF_TRAITS
1
NUMBER_OF_EFFECTS
3
OBSERVATION(S)
5
WEIGHT(S)

EFFECTS: POSITIONS_IN_DATAFILE NUMBER_OF_LEVELS TYPE_OF_EFFECT  [EFFECT NESTED]
4  2 cross
1 15 cross
3  5 cross    # for maternal environmental effect
RANDOM_RESIDUAL VALUES
65.0
RANDOM_GROUP  # first random effect
2
RANDOM_TYPE
add_animal
FILE
mr52.ped
(CO)VARIANCES
20.0
RANDOM_GROUP  # second random effect
3
RANDOM_TYPE
diagonal
FILE

(CO)VARIANCES
15.0
OPTION solv_method FSPAK

Solutions

The results are identical to those in the textbook.

Back to index.html.