[R-sig-ME] Non-convergence error for GLMM with LME4?

[Leanna Jones]

From: leanna_jones at hotmail.com
To: r-help at r-project.org
Subject: Non-convergence error for GLMM with LME4?
Date: Sat, 23 Mar 2013 13:47:47 -0700

Hello!  I am trying to run a GLMM using LME4, and keep getting the warning message: "In mer_finalize(ans) : false convergence (8)"  I am quite new to R, and in looking into this thus far, it appears that there are a variety of reasons why this might occur, such as needing to standardize some parameters or if all subjects in one combination of parameters all have the same outcome.  I also understand that the warning does not necessarily mean that the model results are invalid, but they might be...however, I am unsure how to interpret this in my own situation.

I started with a somewhat more complex model, but kept simplifying it to see if I could get the warning to go away (so it might indicate which predictor variable was the problem...). However, even when using a single fixed effect variable (just sex, for instance), I continue to have the problem, which makes me think the issue may be with my random effect.  Here is the model I would like to run:

mm1=lmer(BinomialOutcome~AgeGroup+Sex+Study.Site+(1|BearID.reformatted),family=binomial)

The study is based on bear captures over a period of time, such that some bears are captured only once, while others many times (in a very unbalanced fashion); I would like to use all the data, but want to account for resampling of specific individuals.  However, this means that there are nearly 600 different bear IDs, and I am wondering if this is the reason why the model will not converge?  If so, what is the best way to address this?  Or other ideas as to what might be going on?

When I run the line above with verbose on, I get the following: 
 > mm1=lmer(BinomialOutcome~AgeGroup+Sex+Study.Site+(1|BearID.reformatted),family=binomial)    #random effect: Ind with intercept of 1, correcting for the intercept/variation btw individuals
  0:     805.13676:  1.12717 -0.556586  1.85210  2.38794  2.36938  2.35604 -2.22337  1.30764  1.34134 -2.47444 -1.63320
  1:     748.91032:  2.05259 -0.834683  1.81369  2.42567  2.40069  2.38217 -2.31535  1.22744  1.26468 -2.64606 -1.74027
  2:     723.97734:  2.74220 -1.05673  2.00374  2.71876  2.57376  2.50128 -2.73578  1.23505  1.35678 -2.98235 -1.63736
  3:     715.51275:  2.92015 -1.58830  2.15652  3.03669  2.81316  2.67649 -3.28992  1.30372  1.36311 -3.32467 -1.84785
  4:     696.36424:  3.21244 -1.51265  2.41407  3.41250  3.22182  2.99137 -3.54075  1.68499  1.83767 -3.33953 -1.80989
  5:     672.69327:  4.20847 -2.35950  3.26463  4.32168  4.39792  4.00460 -3.72829  1.41841  2.56574 -4.27911 -1.86318
  6:     664.48332:  4.37811 -2.58726  3.01153  4.60841  4.65061  4.30175 -4.11225  2.30181  2.55998 -4.38222 -2.02300
  7:     625.68427:  7.31164 -4.89788  5.63003  7.04671  7.95742  8.12903 -4.86511  3.64151  3.86169 -6.33864 -2.89485
  8:     619.38278:  9.39340 -7.21835  5.63590  11.1393  10.6331  11.5975 -4.65786  3.81066  6.58406 -9.13652 -2.96142
  9:     613.14410:  9.53164 -7.17298  6.01739  10.8896  10.7209  11.6709 -4.81531  4.06987  6.53556 -9.01013 -3.05630
 10:     600.31130:  10.0557 -7.29631  7.68818  10.1017  11.1766  12.1945 -5.40073  5.30987  6.73629 -8.62503 -3.67308
 11:     589.17852:  11.9972 -9.32646  9.77915  13.1274  13.2796  14.9828 -6.09763  6.80325  9.19492 -9.58777 -5.25910
 12:     586.76154:  12.3176 -9.61034  9.97050  13.3227  13.7241  15.3584 -6.13498  7.19387  9.59990 -9.83274 -5.36590
 13:     586.09051:  12.3760 -9.59270  9.95463  13.2692  13.7816  15.3696 -6.14017  7.35785  9.64404 -9.83450 -5.34840
 14:     585.97813:  12.3886 -9.59431  9.95680  13.2579  13.7946  15.3724 -6.14235  7.38957  9.65379 -9.83525 -5.34966
 15:     585.77123:  12.4143 -9.59929  9.96353  13.2359  13.8209  15.3784 -6.14715  7.45208  9.67407 -9.83681 -5.35395
 16:     585.69436:  12.4251 -9.60281  9.96832  13.2277  13.8318  15.3810 -6.14945  7.47605  9.68290 -9.83747 -5.35717
 17:     585.54933:  12.4469 -9.61094  9.97946  13.2122  13.8536  15.3863 -6.15434  7.52306  9.70121 -9.83879 -5.36474
 18:     585.49415:  12.4558 -9.61507  9.98519  13.2068  13.8624  15.3886 -6.15653  7.54097  9.70908 -9.83931 -5.36868
 19:     585.38801:  12.4740 -9.62396  9.99753  13.1967  13.8800  15.3933 -6.16106  7.57607  9.72529 -9.84034 -5.37720
 20:     585.38387:  12.4748 -9.62436  9.99809  13.1963  13.8807  15.3935 -6.16126  7.57741  9.72597 -9.84038 -5.37759
 21:     585.35100:  12.4807 -9.62760  10.0026  13.1936  13.8863  15.3951 -6.16281  7.58812  9.73146 -9.84070 -5.38074
 22:     585.35042:  12.4807 -9.62760  10.0026  13.1936  13.8863  15.3951 -6.16281  7.58812  9.73146 -9.87324 -5.38074
 23:     585.22373:  12.5045 -9.64059  10.0207  13.1830  13.9088  15.4016 -6.16903  7.63092  9.75346 -9.87324 -5.39333
 24:     585.21879:  12.5055 -9.64113  10.0216  13.1827  13.9097  15.4018 -6.16930  7.63249  9.75441 -9.87323 -5.39392
 25:     585.20893:  12.5075 -9.64224  10.0233  13.1822  13.9114  15.4024 -6.16984  7.63564  9.75632 -9.87322 -5.39512
 26:     585.20499:  12.5083 -9.64268  10.0240  13.1820  13.9121  15.4026 -6.17006  7.63689  9.75709 -9.87321 -5.39560
 27:     585.20420:  12.5085 -9.64277  10.0241  13.1820  13.9123  15.4027 -6.17011  7.63714  9.75724 -9.87321 -5.39570
 28:     585.19792:  12.5098 -9.64349  10.0252  13.1817  13.9134  15.4030 -6.17046  7.63914  9.75847 -9.87320 -5.39647
 29:     585.19767:  12.5098 -9.64351  10.0253  13.1817  13.9134  15.4030 -6.17047  7.63922  9.75852 -9.87320 -5.39650
 30:     585.19762:  12.5098 -9.64352  10.0253  13.1817  13.9135  15.4030 -6.17047  7.63923  9.75853 -9.87320 -5.39651
 31:     585.19752:  12.5099 -9.64353  10.0253  13.1817  13.9135  15.4030 -6.17048  7.63926  9.75855 -9.87320 -5.39652
 32:     585.19748:  12.5099 -9.64354  10.0253  13.1817  13.9135  15.4030 -6.17048  7.63928  9.75856 -9.87320 -5.39653
 33:     585.19740:  12.5099 -9.64355  10.0253  13.1817  13.9135  15.4031 -6.17049  7.63930  9.75857 -9.87320 -5.39654
 34:     585.19736:  12.5099 -9.64355  10.0253  13.1817  13.9135  15.4031 -6.17049  7.63931  9.75858 -9.87320 -5.39654
 35:     585.19730:  12.5099 -9.64356  10.0253  13.1817  13.9135  15.4031 -6.17049  7.63933  9.75859 -9.87320 -5.39655
 36:     585.19727:  12.5099 -9.64356  10.0253  13.1817  13.9135  15.4031 -6.17049  7.63934  9.75860 -9.87320 -5.39655
 37:     585.19722:  12.5099 -9.64356  10.0253  13.1817  13.9135  15.4031 -6.17050  7.63936  9.75861 -9.87320 -5.39656
 38:     585.19719:  12.5099 -9.64363  10.0253  13.1817  13.9135  15.4031 -6.17050  7.63936  9.75861 -9.87329 -5.39656
 39:     585.19679:  12.5100 -9.64367  10.0254  13.1817  13.9136  15.4031 -6.17052  7.63949  9.75869 -9.87326 -5.39661
 40:     585.19677:  12.5100 -9.64367  10.0254  13.1817  13.9136  15.4031 -6.17052  7.63949  9.75869 -9.87326 -5.39661
 41:     585.19677:  12.5100 -9.64367  10.0254  13.1817  13.9136  15.4031 -6.17052  7.63949  9.75869 -9.87326 -5.39661
 42:     585.19677:  12.5100 -9.64367  10.0254  13.1817  13.9136  15.4031 -6.17052  7.63949  9.75870 -9.87326 -5.39661
 43:     585.19676:  12.5100 -9.64367  10.0254  13.1817  13.9136  15.4031 -6.17052  7.63949  9.75870 -9.87326 -5.39661
 44:     585.19676:  12.5100 -9.64367  10.0254  13.1817  13.9136  15.4031 -6.17052  7.63949  9.75870 -9.87326 -5.39661
 45:     585.19676:  12.5100 -9.64367  10.0254  13.1817  13.9136  15.4031 -6.17052  7.63949  9.75870 -9.87326 -5.39661
 46:     585.19676:  12.5100 -9.64367  10.0254  13.1817  13.9136  15.4031 -6.17052  7.63949  9.75870 -9.87326 -5.39661
 47:     585.19676:  12.5100 -9.64367  10.0254  13.1817  13.9136  15.4031 -6.17052  7.63949  9.75870 -9.87326 -5.39661
 48:     585.19676:  12.5100 -9.64367  10.0254  13.1817  13.9136  15.4031 -6.17052  7.63949  9.75870 -9.87326 -5.39661
 49:     585.19676:  12.5100 -9.64367  10.0254  13.1817  13.9136  15.4031 -6.17052  7.63949  9.75870 -9.87326 -5.39661
 50:     585.19676:  12.5100 -9.64367  10.0254  13.1817  13.9136  15.4031 -6.17052  7.63949  9.75870 -9.87326 -5.39661
 51:     585.19676:  12.5100 -9.64367  10.0254  13.1817  13.9136  15.4031 -6.17052  7.63949  9.75870 -9.87326 -5.39661
 52:     585.19676:  12.5100 -9.64367  10.0254  13.1817  13.9136  15.4031 -6.17052  7.63949  9.75870 -9.87326 -5.39661
Warning message:
In mer_finalize(ans) : false convergence (8)

I read something on another thread about high values (like the -9.6) being a problem here, but I do not completely understand how to interpret or address it. The model will run, and the results appear highly significant, but I do not know to what extent they can be trusted...in case it is relevant, I am using the 64-bit version of R 2.15.1.  (I also tried running it in the 32-bit version, with the same outcome.)  On another thread I also saw a suggestion about maybe trying a development version of R with updates using lme4a(?), but I am not sure where/how I would access that.  Would this be similar to using the pre-release version of R 3.0?

Any help is greatly appreciated!  Thank you!!