Fix issue #896

[rudeboybert]

Attempt at fixing #896. Here is the proposed behavior of augment.lm()

library(tidyverse)
library(broom)
 
# Prep mtcars
mtcars <- mtcars %>%
  rownames_to_column(var = "automobile") %>%
  mutate(cyl = as.factor(cyl)) %>% 
  select(mpg, hp, cyl)

# Fit lm model with y = mpg
mpg_model <- lm(mpg ~ hp + cyl, data = mtcars)

# Use newdata. Output has .resid variable
mtcars %>% 
  augment(mpg_model, newdata = .)
#> # A tibble: 32 x 5
#>      mpg    hp cyl   .fitted  .resid
#>    <dbl> <dbl> <fct>   <dbl>   <dbl>
#>  1  21     110 6        20.0  0.962 
#>  2  21     110 6        20.0  0.962 
#>  3  22.8    93 4        26.4 -3.61  
#>  4  21.4   110 6        20.0  1.36  
#>  5  18.7   175 8        15.9  2.78  
#>  6  18.1   105 6        20.2 -2.06  
#>  7  14.3   245 8        14.2  0.0602
#>  8  24.4    62 4        27.2 -2.76  
#>  9  22.8    95 4        26.4 -3.57  
#> 10  19.2   123 6        19.7 -0.526 
#> # … with 22 more rows

# Use newdata without y. Output does not have .resid variable as expected, and
# no longer has warning message from v0.7.0
mtcars %>% 
  select(-mpg) %>% 
  augment(mpg_model, newdata = .)
#> # A tibble: 32 x 3
#>       hp cyl   .fitted
#>    <dbl> <fct>   <dbl>
#>  1   110 6        20.0
#>  2   110 6        20.0
#>  3    93 4        26.4
#>  4   110 6        20.0
#>  5   175 8        15.9
#>  6   105 6        20.2
#>  7   245 8        14.2
#>  8    62 4        27.2
#>  9    95 4        26.4
#> 10   123 6        19.7
#> # … with 22 more rows

^{Created on 2020-07-13 by the reprex package (v0.3.0)}

[simonpcouch]

Looks great, thanks!

[alexpghayes]

I'll try to find some time for this tonight but I think this approach may be brittle and I'd like to do something else here.

…

On Mon, Jul 13, 2020, 2:34 PM Simon P. Couch ***@***.***> wrote: Merged #897 <#897> into master. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#897 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADTBG26QHDEUAJ4HTB5N5PTR3NOVRANCNFSM4OYWKAZQ> .

[rudeboybert]

I figured. Please do keep me in the loop. I'm curious to see how the pros do it!

[github-actions]

This pull request has been automatically locked. If you believe you have found a related problem, please file a new issue (with a reprex: https://reprex.tidyverse.org) and link to this issue.