BLUPF90

Gibbs sampler for threshold-linear mixed models =====Summary===== Based on THRGIBBSF90 written by DeukHwan Lee at the University of Georgia, 2001 based on GIBBS2F90. Rewritten by Shogo Tsuruta, 2004. Last modified 09/05/2007 =====Functionality===== THRGIBBS1F90 implements Gibbs sampler for mixed threshold-linear models involving multiple categorical and linear variables. Thresholds and variances can be computed or assumed. Another version of thrgibbs1f90b for binary responses is available. The followong options are available: =====Parameters===== The parameter file is the same as for BLUPF90 except for options. =====Options===== <file> OPTION cat 0 0 2 5 </file> 0s indicate that the first and second traits are linear. 2 and 5 indicate that the third and fourth traits are categorical with 2 and 5 categories. <file> OPTION fixed_var all </file> Store all samples for solutions in “all_solutions” and posterior means and SD for all effects in “final_solutions” assuming that (co)variances set in the parameter file are known. <file> OPTION fixed_var all 1 2 3 Store all samples for solutions in “all_solutions” and posterior means and SD for 1, 2, and 3 effects in “final_solutions”. </file> <file> OPTION fixed_var mean </file> Only posterior means and SD for solutions are calculated for all effects in “final_solutions”. <file> OPTION fixed_var mean 1 2 3 Only posterior means and SD for solutions are calculated for effects 1, 2, and 3 in “final_solutions”. </file> <file> OPTION solution all </file> Store all samples for solutions in “all_solutions” and posterior means and SD for all effects. <file> OPTION solution all 1 2 3 </file> Store all samples for solutions in “all_solutions” and posterior means and SD for 1, 2, and 3 effects. <file> OPTION solution mean </file> Only posterior means and SD for solutions are calculated for all effects in “final_solutions”. <file> OPTION solution mean 1 2 3 </file> Only posterior means and SD for solutions are calculated for effects 1, 2, and 3 in “final_solutions”. <file> OPTION cont 10000 </file> 10000 is the number of samples run previously. The user can restart the program from the last run. <file> OPTION prior 10 </file> 10 is the degree of belief for the priors specified in the parameter file. <file> OPTION seed 123 -432 </file> Two seeds for a random number generator can be specified. <file> OPTION thresholds 0.0 1.0 2.0 </file> Set the fixed the thresholds No need to set 0 for binary traits <file> OPTION residual 1 </file> Set the residual variance = 1 <file> OPTION censored xx </file> Negative values of the last category in the data set indicate censored records. “xx” determines the lower and upper limit of the category + xx when sampling from the distribution. <file> OPTION missing -1 </file> Change the missing value for a continuous trait (default 0). This is not tested yet.