2

I'm trying to separate the terms column into two columns the variable used in the regression and the vale of the category. 

  library(MASS)
#> Warning: package 'MASS' was built under R version 3.5.1
  library(dplyr)
#> Warning: package 'dplyr' was built under R version 3.5.1
#> 
#> Attaching package: 'dplyr'
#> The following object is masked from 'package:MASS':
#> 
#>     select
#> The following objects are masked from 'package:stats':
#> 
#>     filter, lag
#> The following objects are masked from 'package:base':
#> 
#>     intersect, setdiff, setequal, union
  library(broom)
#> Warning: package 'broom' was built under R version 3.5.1
as_tibble(Titanic) %>%  dplyr::mutate(y_n = if_else(Survived == "Yes", 1, 0)) %>% 
  glm(y_n ~ Class + n + Age + Sex, data = .) %>%  broom::tidy() %>%  print(n = 10)
#> # A tibble: 7 x 5
#>   term        estimate std.error statistic p.value
#>   <chr>          <dbl>     <dbl>     <dbl>   <dbl>
#> 1 (Intercept)  0.567    0.245       2.31    0.0294
#> 2 Class2nd    -0.00528  0.276      -0.0192  0.985 
#> 3 Class3rd     0.0503   0.279       0.180   0.858 
#> 4 ClassCrew    0.0740   0.283       0.262   0.796 
#> 5 n           -0.00106  0.000907   -1.16    0.255 
#> 6 AgeChild    -0.131    0.225      -0.582   0.566 
#> 7 SexMale      0.0833   0.208       0.401   0.692

Created on 2018-11-02 by the reprex package (v0.2.1)

Something like this is desired:

enter image description here

Ryan John
  • 1,410
  • 1
  • 15
  • 23

1 Answers1

3

Perhaps the following suffices:

df <- as_tibble(Titanic) %>%  dplyr::mutate(y_n = if_else(Survived == "Yes", 1, 0))
m <- glm(y_n ~ Class + n + Age + Sex, data = df)

(trm <- attr(m$terms, "term.labels")) # Getting original variables
# [1] "Class" "n"     "Age"   "Sex"  
(asgn <- attr(model.matrix(m$formula, data = df), "assign")) # See ?model.matrix
# [1] 0 1 1 1 2 3 4 

cbind(Term = trm[asgn[-1]], 
      Category = str_replace(names(coef(m)[-1]), trm[asgn[-1]], ""))
#      Term    Category
# [1,] "Class" "2nd"   
# [2,] "Class" "3rd"   
# [3,] "Class" "Crew"  
# [4,] "n"     ""      
# [5,] "Age"   "Child" 
# [6,] "Sex"   "Male"

The intercept line is missing but, if needed, you may add it in cases when asgn[1] == 0.

Julius Vainora
  • 47,421
  • 9
  • 90
  • 102