MTGSAM: Threshold version
This set of programs is described in "Bayesian analysis of twinning and ovulation rates using a multiple-trait threshold model
and Gibbs sampling" by C.P. Van Tassell, L.D. Van Vleck, and K.E. Gregory (Journal of Animal Science 76:2048–2061, 1998).
Be warned that there are still some bugs in this software. The same problem exists with missing
traits as with the first linear model version (that correction has not yet been applied
to the threshold programs).
Starting values for thresholds for traits with more than 2 categories sometimes blow up when
starting. I have had trouble replicating this problem – if you have a small data set that can generate
this problem, I would appreciate your contacting me and allowing me to use that data for debugging
purposes. Note that if the program does not stop because of this bug, the results should be correct.
I would appreciate any feedback or notice of any bugs or problems encountered with this software.
If you have not done so, please register as a user of MTGSAM!
Please direct questions, comments, and program problems to Curt Van Tassell.
Last Modified: 01/12/2017