> it has converged sufficient for my purposes. > has not converged according to the criterian written into the program, > 5) I do not just increase MAXIT unless I can see the job is converging > 4) XFA2 and XFA3 models are also sometimes slow to converge. > the constrained REML solution, but this is notoriously slow. Then ASReml will use EM steps to approach > of the REML likelihood function occurs at a point outside the > structures when they are constrained Positive Definite, but the maximum > 3) Failure to converge quickly occurs sometimes with US variance This can be recognised by looking at the parameter history > b) Occasionally it is a pair of terms and the parameter estimates are > Typically it is a very small component which is effectively zero. > greater than zero in the variance component table of the. > a) I look for which parameter has not converged: it will have % value > So, When I see that message, I do two things: > parameters, some have changed more than 1 %. > convergence criterion' but when ASReml looks at the % change of the > the likelihood value is changing very slightly (less than the > 2) The message 'LogL Converged, Parameters Not converged' arises when > CONTINUE can allow those PSI to be formally fixed at zero which may help > the fact that for some XFA models with PSI going to the boundary, > starts the second run from where the first finished.
> However, with !CONTINUE, ASReml reads back the.
> run will usually be identical to the first run (except the STEPSIZE > otherwise leave the job unchanged, the first 100 iterations of the new > 1) Having run 100 iterations, if you increase MAXIT to 300 but > Subject: Re: LogL Converged, Parameter Not Converged > From: ASReml users discussion group On Behalf Of Arthur Gilmour > I am still confused with the fact that the model parameters converged with 300 iterations and not with 500 iterations. > For this dataset the model parameters converged with the !CONTINUE qualifier, as confirmed by Dr Cullis. > Source Model terms Gamma Component Comp/SE % C
> How long does ASREML continue to iterate with the !CONTINUE qualifier? The % change for one of the parameters was -1 as shown below On Mon, at 09:55 -0500, KRISHNAMURTHY, SUBASH So a small change can represent a largish %change. It seems to be 2 reps of 26 treatmentsĭivided into 2 groups, in which case it would be fairly orthogonal.Īnd there is no other random term for REP.GROUP1 to oscillate with,īut it is a case where the variance component is small (Comp/SE is. I am surprised that the example below would take more than 300 MAXIT parameter so there is a small chance that the iteration sequence The default value of the stepsize parameter does depend slightly on the I would suspect it stopped because it reached 300, in which case if you However, if you had !MAXIT 300 and it stopped after 300 iterations, Had !MAXIT 500 if that was the ONLY change to the code. Iterations, then I would expect it would stop at the same point if you If you had !MAXIT 300 and iteration stopped at some point before 300 !CONTINUE means CONTINUE from the point were the last run terminatedīy retrieving the variance parameters held in the.