How to plot a QQ plot with 2 samples in ggplot2?

Question

I'd like to plot a QQ plot with x as one sample and y as another (not the common ones which x-axis=theoretical and y-axis=sample) using ggplot2. Does anyone know how to do this?

The common one would be, for example:

ggplot(df, aes(sample=sample)) + stat_qq() 

I am able to do this with qqplot() function,

qqplot(sample1,sample2)

but it seems cannot display multiple plots with grid.arrange(). Or maybe anyone knows some other ways to plot multiple sub-qqplot at one plot?

Answer 1

It's not clear if this is what you mean, but you can add 2 stat_qq calls on the same plot:

library(ggplot2)

set.seed(69)

sample1 <- rnorm(100)
sample2 <- rnorm(100)

ggplot() + 
  stat_qq(aes(sample = sample1), colour = "green") + 
  stat_qq(aes(sample = sample2), colour = "red") +
  geom_abline(aes(slope = 1, intercept = 0), linetype = 2)

The other possibility is that you mean to have one sample on the x axis and one sample on the y axis to compare their relative ordering. You could do that like this:

ggplot(mapping = aes(x = sort(sample1), y = sort(sample2))) + 
  geom_point() +
  geom_abline(aes(slope = 1, intercept = 0), linetype = 2)

---

Edit

From the comments, it is clear the OP is looking for an empirical qq plot of two vectors. This function should provide a rough ggplot equivalent of the function referenced in the comments:

gg_qq_empirical <- function(a, b, quantiles = seq(0, 1, 0.01))
{
  a_lab <- deparse(substitute(a))
  if(missing(b)) {
    b <- rnorm(length(a), mean(a), sd(a))
    b_lab <- "normal distribution"
  }
  else b_lab <- deparse(substitute(b))
  
  ggplot(mapping = aes(x = quantile(a, quantiles), 
                       y = quantile(b, quantiles))) + 
    geom_point() +
    geom_abline(aes(slope = 1, intercept = 0), linetype = 2) +
    labs(x = paste(deparse(substitute(a)), "quantiles"), 
         y = paste(deparse(substitute(b)), "quantiles"),
         title = paste("Empirical qq plot of", a_lab, "against", b_lab))
}

So, for example we can do:

qq <- gg_qq_empirical(sample1, sample2)
qq + theme_light() + coord_equal()

^{Created on 2020-07-03 by the reprex package (v0.3.0)}

Answer 2

d <- tibble(Group=rep(1:2, each=200), Sample=c(rnorm(200), rnorm(200, mean=2, sd=4)))
d %>% ggplot(aes(sample=Sample)) + stat_qq() + facet_wrap(~Group)
qplot(sample=Sample, data=d, color=as.factor(Group))

The stat_qq() call gives

The qplot() call gives