[R-sig-ME] singular convergence with lmer()

[David Duffy]

On Sun, 8 Jul 2012, Reinhold Kliegl wrote:

> It converged for me for lme4_0.999999-0.
> Estimates look different from what you posted at the site.
> Reinhold Kliegl
>
>> dat$Part <- factor(dat$Part)
>> ( fit <- lmer(y ~ (1|Operator)+(1|Part)+(1|Part:Operator), data=dat) )
> Linear mixed model fit by REML
> Formula: y ~ (1 | Operator) + (1 | Part) + (1 | Part:Operator)
>   Data: dat
>    AIC    BIC logLik deviance REMLdev
> -619.7 -603.4  314.9   -630.3  -629.7
> Random effects:
> Groups        Name        Variance   Std.Dev.
> Part:Operator (Intercept) 0.00081854 0.028610
> Part          (Intercept) 1.06721729 1.033062
> Operator      (Intercept) 0.00031226 0.017671
> Residual                  0.00063295 0.025159
> Number of obs: 192, groups: Part:Operator, 96; Part 12; Operator, 8
>
> Fixed effects:
>            Estimate Std. Error t value
> (Intercept)   2.7171     0.2983   9.109
>

The regress package (1.3-8) also gives

  r1 <- regress(y ~ 1, ~ Operator+Part+I(Operator:Part), data=dat)
  r1$sigma
         Operator             Part I(Operator:Part)               In
     0.0003122871     1.0671419239     0.0008185268     0.0006329560