Echidna now sets variances in a diag() structure which approach zero to exactly zero which is equivalent to dropping the associated effects from the model. Be wary in case this new feature doesn’t work properly in your particular model.Several issues with fitting Factor Analytic models […]
Patches and recent notes
Allow at(Loc,’Orange’,’Bathurst’).Geno model term syntax. The class labels may not be truncated.!RES qualifier saves data and residuals in a _RES.bin binary file.Some bug fixesAbsorb DIAG-NODE improves speed in some jobsSparse Ci MTNODE allows GRM models to run fasterSystemEnvironment variable Echidna_qual=… parsed after the data line. […]
Revised parsing of the model line to resolve a few issues. Hopefully everything else still works.Revised processing of !A and !I class name lists to resolve some problems. Hopefully all issues are now resolved. If you have build 1.51, I recommend you update it promptly.Allowed […]
Fix bug reading large data with many missing valuesFix bug calculating score for bivariate, unbalanced data and singularities
Fix bug with id(units).us(Tr) processingFix bug processing predict gen when gen modelled via Cgen (a contrast)Faster prediction. The predict absorption step has been reorganised and now runs substantially faster but let me know of any issues.
Fix bug absorbing working variables in a large jobFix bug in control of sorting data for spatial analysis Detect error in the pedigree fileAllow for longer lines in data file (previous limit was 5000 characters)Fix bugs encoding factors when (un)declared initial size is not enough
Fix bug absorbing working variables in a large jobFix bug in control of sorting of data recordsDetect error in the pedigree file
Added check for large residualsAdded calc of slope of log(abs(residual)) vs log(pv)Added a green Normal Probablity plot to top left corner of basic residual plot.Implemented !LAST functionality for Fixed Pedigree groupsAutomatic !CSVSTYLE if pedigree file is .csv
Cosmetic changes to .esr file layoutBetter (I hope) labelling of levels in .ess file At Easter time many remember Jesus was executed for our sins at Passover AD30so that anyone who believes He rose and trusts Him will participate in His Kingdom.
Activated bivariate analysis for non-normal (GLMM) traits. Bivariate GLMM (March 2021) When specifying a bivariate analysis where 1 or both traits are non normal, the distribution and link qualifiers must follow the variable they relate to. For example Sick !BINOMIAL !LOGIT LiveWt ~ Trait Trait.Sex […]