CODEWITHSUNDEEP
rloopsggplot2correlation
perherring
perherring

Reputation: 21

Using Loops to create multiple correlation scatterplots

I am looking to generate ggplot2 scatter plots for each variable combination.

We can use the mtcars dataset as an example -- how can I generate the following graph for every possible column combination?

mpg vs. cyl, mpg vs. disp, etc...

cyl vs. disp, cyl vs. hp, etc...

and so on....

ggplot(mtcars, aes(x = mpg, y = cyl)) +
  geom_point(size = 3.5) +
  guides(fill=guide_legend(override.aes=list(shape=21))) +
  stat_cor(method = "pearson") +
  geom_smooth(method = lm, alpha = 0.1, color = "black")

can a loop be generated and column names be recorded and graphs are sequentially generated and saved as a .tif?

Upvotes: 0

Views: 45

Answers (1)

Wael
Wael

Reputation: 1800

If your aim is to have var to var correlation plots there are some packages for that like GGally

library(GGally)
ggscatmat(mtcars,columns =c(1,3:5) ,  color = "vs", alpha = 0.8)

Created on 2023-04-28 with reprex v2.0.2

but if your aim is to use your customized plot

library(ggplot2)
library(ggpubr)
for (i in names(mtcars)[1:3]) {
  for (j in names(mtcars)[1:3]) {
    g=ggplot(mtcars, aes(x =!!sym(i), y = !!sym(j))) +
      geom_point(size = 3.5) +
      guides(fill=guide_legend(override.aes=list(shape=21))) +
      stat_cor(method = "pearson") +
      geom_smooth(method = lm, alpha = 0.1, color = "black")
    print(g)
  }
}

.... ....

Created on 2023-04-28 with reprex v2.0.2

Upvotes: 0

Related Questions