Geek Answers Handbook

R 3.3.2: lme4 + lmerTest problems on Mac OS Sierra

I came across a problem that affects Mac OS R version 3.3.2 (and .3 too!) When using lme4 and lmerTest.

lmerTest creates an error:

Error calculating the Satterthwaite approximation. The output of the lme4 package is returned. A summary from lme4 is displayed in lmerTest Some computational error

The problem does not emerge with R 3.2 under MacOS and any R versions under Windows. However, this is not an installation problem, since I reproduced the error after reinstalling R and also on another Mac.

Here is a sample code:

library(lme4) #' start of data creation mydat <- structure(list(ID = c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 27, 28, 29, 1, 2, 3, 4, 5, 6, 7, 8, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29), sex = c(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1), ROI = structure(c(4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L), .Label = c("calf", "DSCAT", "KM", "neck", "SSCAT", "VAT"), class = "factor"), value = c(0.674, 0.561, 0.543, 0.563, 0.697, 0.608, 0.56, 0.448, 0.626, 0.515, 0.568, 0.528, 0.587, 0.532, 0.547, 0.514, 0.587, 0.572, 0.559, 0.569, 0.462, 0.531, 0.477, 0.582, 0.583, 0.569, 0.563, 0.576, 0.84, 0.638, 0.69, 0.707, 0.704, 0.627, 0.769, 0.637, 0.515, 0.669, 0.699, 0.626, 0.59, 0.639, 0.501, 0.632, 0.624, 0.641, 0.669, 0.656, 0.556, 0.569, 0.633, 0.608, 0.616, 0.664, 0.666, 0.669, 0.545, 0.514, 0.45, 0.585, 0.547, 0.572, 0.577, 0.458, 0.47, 0.537, 0.532, 0.455, 0.62, 0.501, 0.506, 0.44, 0.499, 0.577, 0.457, 0.481, 0.522, 0.516, 0.513, 0.559, 0.571, 0.515, 0.575, 0.521, 0.44, 0.637, 0.521, 0.634, 0.552, 0.581, 0.55, 0.553, 0.522, 0.634, 0.631, 0.512, 0.603, 0.593, 0.58, 0.442, 0.53, 0.463, 0.587, 0.538, 0.48, 0.557, 0.482, 0.53, 0.592, 0.445, 0.526, 0.45, 0.551, 0.51, 0.678, 0.64, 0.599, 0.589, 0.627, 0.621, 0.601, 0.526, 0.619, 0.599, 0.668, 0.615, 0.621, 0.561, 0.532, 0.56, 0.578, 0.686, 0.57, 0.457, 0.563, 0.61, 0.513, 0.638, 0.594, 0.777, 0.562, 0.663, 0.538, 0.471, 0.518, 0.47, 0.535, 0.644, 0.605, 0.474, 0.468, 0.563, 0.539, 0.47, 0.538, 0.453, 0.494, 0.576, 0.418, 0.609, 0.528, 0.453, 0.569, 0.484, 0.486, 0.558, 0.621, 0.465, 0.691, 0.398, 0.539, 0.574), Alter = c(45, 47, 51, 44, 35, 26, 60, 44, 42, 50, 42, 51, 57, 23, 26, 29, 29, 50, 45, 61, 61, 58, 32, 27, 49, 45, 64, 28, 45, 47, 51, 44, 35, 26, 60, 44, 50, 42, 51, 57, 23, 26, 29, 29, 50, 45, 61, 61, 58, 32, 27, 49, 27, 45, 64, 28, 45, 47, 51, 44, 35, 26, 60, 44, 42, 50, 42, 51, 57, 23, 26, 29, 29, 50, 45, 61, 61, 58, 32, 27, 49, 27, 45, 64, 28, 45, 47, 51, 44, 35, 26, 60, 44, 42, 50, 42, 51, 57, 23, 26, 29, 29, 50, 45, 61, 61, 58, 32, 27, 49, 27, 45, 64, 28, 45, 47, 51, 44, 35, 26, 60, 44, 42, 50, 42, 51, 57, 23, 26, 29, 29, 50, 45, 61, 61, 58, 32, 27, 49, 27, 45, 64, 28, 45, 47, 51, 44, 35, 26, 60, 44, 42, 50, 42, 51, 57, 23, 26, 29, 29, 50, 45, 61, 61, 58, 32, 27, 49, 27, 45, 64, 28), BMI = c(29.7506923675537, 28.8, 28.8385677337646, 41.48, 27.7186069488525, 29.54, 38.06, 35.8453826904297, 35.57, 31.77, 31.75, 32.78, 30.5336246490479, 29.1074104309082, 36.4690246582031, 31.7769088745117, 31.5393238067627, 31.5596752166748, 27.593786239624, 30.8192825317383, 27.0799140930176, 31.481481552124, 29.0328979492188, 24.52, 29.4029197692871, 35.6112785339355, 28.2401905059814, 28.8979587554932, 29.7506923675537, 28.8, 28.8385677337646, 41.48, 27.7186069488525, 29.54, 38.06, 35.8453826904297, 31.77, 31.75, 32.78, 30.5336246490479, 29.1074104309082, 36.4690246582031, 31.7769088745117, 31.5393238067627, 31.5596752166748, 27.593786239624, 30.8192825317383, 27.0799140930176, 31.481481552124, 29.0328979492188, 24.52, 29.4029197692871, 23.0956573486328, 35.6112785339355, 28.2401905059814, 28.8979587554932, 29.7506923675537, 28.8, 28.8385677337646, 41.48, 27.7186069488525, 29.54, 38.06, 35.8453826904297, 35.57, 31.77, 31.75, 32.78, 30.5336246490479, 29.1074104309082, 36.4690246582031, 31.7769088745117, 31.5393238067627, 31.5596752166748, 27.593786239624, 30.8192825317383, 27.0799140930176, 31.481481552124, 29.0328979492188, 24.52, 29.4029197692871, 23.0956573486328, 35.6112785339355, 28.2401905059814, 28.8979587554932, 29.7506923675537, 28.8, 28.8385677337646, 41.48, 27.7186069488525, 29.54, 38.06, 35.8453826904297, 35.57, 31.77, 31.75, 32.78, 30.5336246490479, 29.1074104309082, 36.4690246582031, 31.7769088745117, 31.5393238067627, 31.5596752166748, 27.593786239624, 30.8192825317383, 27.0799140930176, 31.481481552124, 29.0328979492188, 24.52, 29.4029197692871, 23.0956573486328, 35.6112785339355, 28.2401905059814, 28.8979587554932, 29.7506923675537, 28.8, 28.8385677337646, 41.48, 27.7186069488525, 29.54, 38.06, 35.8453826904297, 35.57, 31.77, 31.75, 32.78, 30.5336246490479, 29.1074104309082, 36.4690246582031, 31.7769088745117, 31.5393238067627, 31.5596752166748, 27.593786239624, 30.8192825317383, 27.0799140930176, 31.481481552124, 29.0328979492188, 24.52, 29.4029197692871, 23.0956573486328, 35.6112785339355, 28.2401905059814, 28.8979587554932, 29.7506923675537, 28.8, 28.8385677337646, 41.48, 27.7186069488525, 29.54, 38.06, 35.8453826904297, 35.57, 31.77, 31.75, 32.78, 30.5336246490479, 29.1074104309082, 36.4690246582031, 31.7769088745117, 31.5393238067627, 31.5596752166748, 27.593786239624, 30.8192825317383, 27.0799140930176, 31.481481552124, 29.0328979492188, 24.52, 29.4029197692871, 23.0956573486328, 35.6112785339355, 28.2401905059814, 28.8979587554932)), .Names = c("ID", "sex", "ROI", "value", "Alter", "BMI"), row.names = c(NA, -172L), class = c("tbl_df","tbl", "data.frame")) #' end of data creation library(lmerTest) mod <- lmer(value~Alter+ROI+BMI+(1|ID),data=mydat,REML=F) summary(mod) sessionInfo()

System information is as follows:

 R version 3.3.3 (2017-03-06) Platform: x86_64-apple-darwin13.4.0 (64-bit) Running under: macOS Sierra 10.12.3 locale: [1] C attached base packages: [1] stats graphics grDevices utils datasets methods base other attached packages: [1] lmerTest_2.0-33 lme4_1.1-12 Matrix_1.2-8 loaded via a namespace (and not attached): [1] Rcpp_0.12.9 Formula_1.2-1 knitr_1.15.1 magrittr_1.5 cluster_2.0.5 splines_3.3.3 MASS_7.3-45 munsell_0.4.3 [9] colorspace_1.3-2 lattice_0.20-34 minqa_1.2.4 stringr_1.1.0 plyr_1.8.4 tools_3.3.3 nnet_7.3-12 grid_3.3.3 [17] data.table_1.10.0 checkmate_1.8.2 htmlTable_1.8 gtable_0.2.0 nlme_3.1-131 latticeExtra_0.6-28 htmltools_0.3.5 digest_0.6.11 [25] survival_2.40-1 lazyeval_0.2.0 assertthat_0.1 tibble_1.2 gridExtra_2.2.1 RColorBrewer_1.1-2 nloptr_1.0.4 ggplot2_2.2.1 [33] base64enc_0.1-3 acepack_1.4.1 rpart_4.1-10 stringi_1.1.2 backports_1.0.4 scales_0.4.1 Hmisc_4.0-2 foreign_0.8-67
+7
r macos-sierra lme4 macos lmertest
source share
2 answers

After retrying, the code worked under R3.3.3, although my system has not changed. I dreamed? Kind of paranormal I ... I'm puzzled. Excuse for troubling.

R version 3.3.3 (2017-03-06) Platform: x86_64-apple-darwin13.4.0 (64-bit) Works under: macOS Sierra 10.12.3

locale: [1] C

attached base packages: [1] grDevices utils graphics statistics
database methods database

other attached packages: [1] lmerTest_2.0-33 lme4_1.1-12
Matrix_1.2-8

loaded through the namespace (and not attached): [1] Rcpp_0.12.9
nloptr_1.0.4 RColorBrewer_1.1-2 plyr_1.8.4
base64enc_0.1-3 tools_3.3.3 rpart_4.1-10
digest_0.6.12 [9] tibble_1.2 nlme_3.1-131
gtable_0.2.0 htmlTable_1.9 checkmate_1.8.2
lattice_0.20-34 gridExtra_2.2.1 stringr_1.2.0 [17] cluster_2.0.5 knitr_1.15.1 htmlwidgets_0.8 grid_3.3.3 nnet_7.3-12 data.table_1.10.0 survival_2.40-1
foreign_0.8-67 [25] latticeExtra_0.6-28 minqa_1.2.4
Formula_1.2-1 ggplot2_2.2.1 magrittr_1.5
Hmisc_4.0-2 scales_0.4.1 backports_1.0.5 [33] htmltools_0.3.5 MASS_7.3-45 splines_3.3.3
assertthat_0.1 colorspace_1.3-2 stringi_1.1.2
acepack_1.4.1 lazyeval_0.2.0 [41] munsell_0.4.3

+1
source share

This is not really an answer, but it is a bit long for comment ...

I cannot reproduce this in any of these environments:

 R version 3.3.2 (2016-10-31) Platform: x86_64-apple-darwin13.4.0 (64-bit) Running under: OS X El Capitan 10.11.6 [1] lmerTest_2.0-33 lme4_1.1-12 Matrix_1.2-8 (also tried with Matrix 1.2-7) R Under development (unstable) (2017-02-13 r72168) Platform: x86_64-pc-linux-gnu (64-bit) Running under: Ubuntu 14.04.5 LTS lmerTest_2.0-33 lme4_1.1-13 Matrix_1.2-8

Without replicability, this is quite difficult to fix. It's a little hard to believe that Sierra-specific, but strange things have happened.

I'm going to make a stupid assumption and suggest that you try to downgrade the Matrix package to version 1.2-7 (as described here ), although both symptoms [crashes] and suspicious platforms [32-bit OS] are different.

Alternatively, you can try digging lmerTest into the guts as described here to find out what happens, although again the specific context is different (your fit model is not singular).

CRAN does check packages under the 64-bit version of Sierra , but for lmerTest (and for lme4 ) no errors are displayed on this platform ...

0
source share

More articles:


All Articles