Patches and recent notes

Now handles herd.ar1v(td) where td has many levels,idv(herd).ar1(td) is equivalent and works.Fixed some bugs when using grouped variables; some functions accessed the wrong data variable

Fixed bugs handling residual sat(env).units.diag(Trait,$Ve) Thanks for continuing to work with Echidna. Thankyou to those who have reported issues for me to fix Best wishes for your journey with the Lord Jesus in 2021. He is near to all who call on Him. Arthur

Fixed bug In cycle mode, accumulate to ESS, ESY, ESX, ETB, EPV, EVC etcFiddle ASReml equation ordering routine (!EQN 7)Fixed Bug in ALPHA labels space allocationFixed Bug reporting some errors I wish you all the best for Christmas and the New Year. Jesus came as […]

Update the ASReml equation ordering routine (!EQN 7)Fix case of xfa(a).b when a has more levels than bCheck the Resources TAB for new material you might like to read.

Various BUG fixes (14 routines) Speed up absorption of working variable cross products when forming AI matrix basic mbf() design function added

Fix scale in Legendre polynomial coefficientsReport potential issue if grid sorting variables are not simple factors

Fix bug reporting PREDICT values for a covariateFaster absorb step in some very large modelsFix conversion of var(loadings) from Gamma to Sigma scale (VPREDICT)Allow at(Rep,0) as alternative to zero and zero(1) in the modelWarn if GRM nearly singular.Bug getting Score for GDENSE Sparse stored GIV

Fix conversion of var(loadings) from Gamma to Sigma scale (VPREDICT)Allow at(Rep,0) as alternative to zero and zero(1) in the modelWarn if GRM nearly singular.

Fixed bug in calculating covariances among variance components estimated on GAMMA scale which was affecting VPREDICT calculations. Thanks to all who report issues they detect, especially comparing results with those from ASReml.

coruh() variance structure implemented