I can run the same regression using different data frames using a pipe and loop (see below):
mtcars %>%
split(.$cyl) %>%
map(~ lm(mpg ~ wt, data = .x))
However, what if I am interested seeing results for the same regression but for a range of different dependent variables - e.g. "mpg", "hp", "drat". Is there a fast way to do this using loops?
I have tried using nested lapply loops, group_by etc. however, I can't seem to find a solution.
Any help would be great.
It sounds like you wish to loop through the column names of each data frame. Effectively you need a double map or double lapply. Something like this would work:
mtcars %>%
split(.$cyl) %>%
lapply(function(x)
{
lapply(paste("mpg ~", names(x)[-1]), function(y) {
lm(formula = as.formula(y), data = x)
})})
#> $`4`
#> $`4`[[1]]
#>
#> Call:
#> lm(formula = as.formula(y), data = x)
#>
#> Coefficients:
#> (Intercept) cyl
#> 26.66 NA
#>
#>
#> $`4`[[2]]
#>
#> Call:
#> lm(formula = as.formula(y), data = x)
#>
#> Coefficients:
#> (Intercept) disp
#> 40.8720 -0.1351
# ... etc (very long list)
The tidy loop could be:
library(tidyverse)
mtcars %>%
select(-mpg) %>%
map(~lm(mtcars$mpg ~ .x), data = mtcars)
This might look like:
library(dplyr)
label <- "mpg"
features <- setdiff(names(mtcars), label)
generate_formula <- function(feature) sprintf("%s ~ %s", label, feature) %>% as.formula
features %>%
map(~ lm(generate_formula(.x), data = mtcars))
Output:
[[1]]
Call:
lm(formula = generate_formula(.x), data = mtcars)
Coefficients:
(Intercept) cyl
37.885 -2.876
[[2]]
Call:
lm(formula = generate_formula(.x), data = mtcars)
Coefficients:
(Intercept) disp
29.59985 -0.04122
[[3]]
Call:
lm(formula = generate_formula(.x), data = mtcars)
Coefficients:
(Intercept) hp
30.09886 -0.06823
We can use nest_by
library(dplyr)
mtcars %>%
nest_by(cyl) %>%
mutate(model = list(lm(mpg~ ., data = data)))
# A tibble: 3 x 3
# Rowwise: cyl
# cyl data model
# <dbl> <list<tbl_df[,10]>> <list>
#1 4 [11 × 10] <lm>
#2 6 [7 × 10] <lm>
#3 8 [14 × 10] <lm>
