how_to_estimate_different_residual_co_variances_by_class_in_airemlf90_as_in_gibbs3f90
This is an old revision of the document!
Basic idea is: to specify dummy covariates (“0” or “1”) in the “hetres_pos” option.
In this example, we will consider 3 different residual variances by class in a single-trait model. We need extra 3 columns in the data file. Each column contains the following information:
extra column 1: put "1" if this observation is related to the residual variance 1; put "0" elsewhere. extra column 2: put "1" if this observation is related to the residual variance 2; put "0" elsewhere. extra column 3: put "1" if this observation is related to the residual variance 3; put "0" elsewhere.
We assume these columns are corresponding to 7-th, 8-th, and 9-th columns in the data file, respectively.
The following lines should be added to your parameter file:
OPTION hetres_pos 7 8 9 OPTION hetres_pol 20.0 20.0 20.0
An initial value should be specified as natural logarithm of the variance. For example, 20.0 means ln(20) = 3.
how_to_estimate_different_residual_co_variances_by_class_in_airemlf90_as_in_gibbs3f90.1447189740.txt.gz · Last modified: 2024/03/25 18:22 (external edit)