More components allowed in VPREDICT Fixed bug (crash( reporting long integer class names in PREDICTAdded ANTE variance structure . ante1(access) as R and GSome changes to STEPSZ and SLOW tuning have slightly modified some convergence sequences. Added ar1h(access) as R and GFixed ante1(access $VarG) where […]
Added PREDICT … !CONTRAST coeffs to facilitate testing treatment contrasts Small Speed up of SCORE calculation in bivariate analysis Fixed bug reporting solutions in 1.61
Fixed bug parsing a long (>200characters) !ASSIGN listFixed bug associated with settng specific variances to zero.Added !NCZ qualifier to !MBF as alternative to setting columns in function 2nd argument. If you are not at least using build 1.60, update now.
Fix recent (1.59) bug in processing with !GDENSE qualifier.Fix recent (1.58) bug reporting labels of class levels in .ess/.etb files From build 1.58, if Echidna is invoked with no arguments, it entersa MenuMode with options to and edit an existing command file,create a command file […]
This build incorporates faster processing of larger jobs achieved by using block processing of the mixed model equation for absorption and inversion. The gain depends on the sparsity pattern of the mixed model equations and is greatest with less sparse patterns.
From build 1.58, if Echidna is invoked with no arguments, it entersa MenuMode with options to and edit an existing command file,create a command file based on a data file (with variable labels)and process the command file. This will allow Mac users to use Finder. […]
Thank God for all His blessings in 2021 through Jesus Christ.At Christmas we remember Jesus came to give his life a ransom for all. Fix intermittent bug reporting class names of PEDIGREE factor.Fix addressing errors with large workspace.
Extended PREDICT … !TDIFF to report critical values when using residual idv(units)
Calculate Denominator DF and F probability in simple univariate modelsReport Bonferroni & Tukey critical values for PREDICT treatment !TDIFFFix XML formatting errors
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 […]