Skip to content

rms::latex.anova broken?

2 messages · Kevin E. Thorpe

Kevin E. Thorpe
# 
I am re-running some logistic regression analyses using lrm from the rms 
package but latex(anova(...)) appears to be broken on my system.

Here is some anova() output followed by the latex() error for two models 
since the error changes. My sessionInfo() follows the other output. I 
have updated all my packages and re-installed Hmisc and rms plus 
dependencies. The only thing I haven't done yet is update R completely. 
Has anyone else encountered this and know how to solve it?

 > anova(full)
                 Wald Statistics          Response: id14

  Factor              Chi-Square d.f. P
  birthweight_kilo     0.87       1   0.3517
  ageinmonth           4.12       1   0.0423
  zbmi                 6.49       1   0.0108
  maxtbf              15.16       1   0.0001
  cowsmilk             6.54       1   0.0106
  Male                 1.96       1   0.1611
  multivitamin         0.76       1   0.3819
  bottleuse            0.13       1   0.7194
  preterm              0.50       1   0.4811
  AGEINTRO_cowsmilk    0.06       1   0.8032
  AGEINTRO_complefood  0.61       1   0.4356
  TOTAL               30.49      11   0.0013
 > latex(anova(full),file="",table.env=FALSE,booktabs=TRUE)
Error in ifelse(sn %nin% c("d.f.", "MS", "Partial SS"), math(sn), sn) :
   could not find function "math"
 > anova(full.nl)
                 Wald Statistics          Response: id14

  Factor              Chi-Square d.f. P
  birthweight_kilo     3.68       2   0.1588
   Nonlinear           2.65       1   0.1037
  ageinmonth          16.25       2   0.0003
   Nonlinear          13.45       1   0.0002
  zbmi                 4.07       2   0.1310
   Nonlinear           0.23       1   0.6323
  maxtbf              15.81       2   0.0004
   Nonlinear           2.57       1   0.1092
  cowsmilk             3.34       2   0.1880
   Nonlinear           1.16       1   0.2821
  Male                 1.21       1   0.2711
  multivitamin         0.57       1   0.4494
  bottleuse            0.06       1   0.8100
  preterm              0.01       1   0.9418
  AGEINTRO_cowsmilk    3.65       2   0.1612
   Nonlinear           3.28       1   0.0700
  AGEINTRO_complefood  5.40       2   0.0671
   Nonlinear           4.00       1   0.0455
  TOTAL NONLINEAR     25.41       7   0.0006
  TOTAL               52.13      18   <.0001
 > latex(anova(full.nl),file="",table.env=FALSE,booktabs=TRUE)
Error in paste0(specs$lspace, specs$italics(substring(rowl, 2)), sep = 
"") :
   attempt to apply non-function

 > sessionInfo()
R version 3.2.3 Patched (2016-01-31 r70055)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Slackware 14.2

locale:
  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C
  [3] LC_TIME=en_US.UTF-8        LC_COLLATE=C
  [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8
  [7] LC_PAPER=en_US.UTF-8       LC_NAME=C
  [9] LC_ADDRESS=C               LC_TELEPHONE=C
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base

other attached packages:
[1] rms_5.1-0       SparseM_1.74    Hmisc_4.0-2     ggplot2_2.2.1
[5] Formula_1.2-1   survival_2.40-1 lattice_0.20-34 knitr_1.15.1

loaded via a namespace (and not attached):
  [1] Rcpp_0.12.9         RColorBrewer_1.1-2  plyr_1.8.4
  [4] base64enc_0.1-3     tools_3.2.3         rpart_4.1-10
  [7] digest_0.6.12       polspline_1.1.12    tibble_1.2
[10] gtable_0.2.0        htmlTable_1.9       checkmate_1.8.2
[13] nlme_3.1-131        Matrix_1.2-8        mvtnorm_1.0-5
[16] gridExtra_2.2.1     stringr_1.1.0       cluster_2.0.5
[19] htmlwidgets_0.8     MatrixModels_0.4-1  grid_3.2.3
[22] nnet_7.3-12         data.table_1.10.4   foreign_0.8-67
[25] multcomp_1.4-6      TH.data_1.0-8       latticeExtra_0.6-28
[28] magrittr_1.5        codetools_0.2-15    MASS_7.3-45
[31] scales_0.4.1        backports_1.0.5     htmltools_0.3.5
[34] splines_3.2.3       assertthat_0.1      colorspace_1.3-2
[37] quantreg_5.29       sandwich_2.3-4      stringi_1.1.2
[40] acepack_1.4.1       lazyeval_0.2.0      munsell_0.4.3
[43] zoo_1.7-14
1 day later
Kevin E. Thorpe
# 
I figured I should follow-up to say the function is not broken, rather 
it was me not paying close attention to recent release notes.

Frank Harrell kindly informed me (thanks Frank) that with recent 
versions of Hmisc/rms I should use options(prType="latex") for LaTeX 
output to behave correctly.

Kevin
On 02/07/2017 02:23 PM, Kevin E. Thorpe wrote: