I have a table called join_boin. and there is 4 column.
list <- c("a", "b", "c", "d)
function <- function(a){
for (i in list){
table %>%
ggplot(aes(!!as.name(i)))+
geom_histogram(aes(y=amount)) +
stat_function(fun = dnorm, args = list(mean = mean(table$a), sd = sd(table$a))) -> result
print(result)
}
}
When I use a loop to create a distribution for 4 variable's histograms, I don't know how to deal with (mean = mean(table $ DIFFERENT VARIABLES) ...
Instead of $, use [[. In the OP's code, the argument passed into function is av_total, whereas the for loop uses the list object created outside the function. It should be the same as av_total (or change the argument to 'av_total_list')
functiona <- function(data, av_total_list){
for(av_total in av_total_list) {
data %>%
ggplot(aes(!!as.name(av_total)))+
geom_histogram(aes(y= ..density..), bins = 42) +
stat_function(fun = dnorm, colour = "red",
args = list(mean = mean(data[[av_total]]),
sd = sd(data[[av_total]]))) -> myresult
print(myresult)
}
}
and call it as
functiona(join_boston, list)
reproducible with
vec <- c('mpg', 'disp', 'hp')
functiona(mtcars, vec)
The above function just print the output into the console. Instead, it may be better to create a function that return the output and loop over the list and apply the function
library(dplyr)
library(ggplot2)
library(purrr)
functionb <- function(data, av_total) {
data %>%
ggplot(aes(.data[[av_total]])) +
geom_histogram(aes(y = ..density..), bins = 42) +
stat_function(fun = dnorm, colour = "red",
args = list(mean = mean(data[[av_total]], na.rm = TRUE),
sd = sd(data[[av_total]], na.rm = TRUE)))
}
and call the function by loop over the column names
out <- map(list, ~ functionb(join_boston, .x))
As a reproducible example for testing
data(mtcars)
library(ggpubr)
vec <- c('mpg', 'disp', 'hp')
out <- map(vec, ~ functionb(mtcars, .x))
ggarrange(plotlist = out, ncol = 2, nrow = 2)
-output
