Use group summary values as geom_ribbon in ggplot2?

Question

I am trying to plot a line plot by groups, where I would like to add underling shading covering min and max values for categories. I would like to replicate plot like this, but grey shade covers all between the dashed and solid black lines (not showing confidence interval):

Normally, this is done using geom_ribbon where I need to specify values for aes(ymin = ymin, ymax = ymax). But how to pass the ymin and ymax values for values of group c and a (in my example)?

Here is my dummy example:

# example for shaded line plot
dd <- data.frame(year = c(1:5),
                 group = rep(c("a", "b", "c"), each = 5),
                 vals = c(5, 5.2, 5.6, 5.8, 6,
                          5, 4.9, 4.8, 4.7, 4.2,
                          5, 4.8, 4.4, 4,   3))
dd %>% 
  ggplot(aes(x = year,
             y = vals,
             shape = group,     
             color = group,     
             linetype = group)) +
  ylim(0,6.5) +
  geom_line(lwd  = 1.5)  +
  theme_bw()

My data are in long format and I have many groups, so I would appreciate some automatic solution. Thank you in advance.

My approach:

convert data from long to wide format to pass argumens as ymin and ymax. Not working as using two different dataframes:

# maybe first convert data from long to wide?
library(tidyr)

dd.spread <- spread(dd, group, vals)

dd %>% 
  ggplot(aes(x = year,
             y = vals,
             shape = group,     
             color = group,     
             linetype = group)) +
  geom_line()  +
  geom_ribbon(dd.spread, aes(ymin = c, ymax = a), color = "grey70")

Expected outcome (shaded between a-c):

score 1 · Accepted Answer · answered Apr 19 '21 at 13:24

You can put in a sneaky pivot_wider at the data argument of a layer. In the example below I migrated some aesthetics from the main plot call to the line layer, as they were conflicting in the ribbon layer.

library(ggplot2)
library(tidyr)
#> Warning: package 'tidyr' was built under R version 4.0.3

# example for shaded line plot
dd <- data.frame(year = c(1:5),
                 group = rep(c("a", "b", "c"), each = 5),
                 vals = c(5, 5.2, 5.6, 5.8, 6,
                          5, 4.9, 4.8, 4.7, 4.2,
                          5, 4.8, 4.4, 4,   3))
dd %>% 
  ggplot(aes(x = year)) +
  ylim(0,6.5) +
  geom_ribbon(
    data = ~ pivot_wider(., names_from = group, values_from = vals),
    aes(ymin = c, ymax = a)
  ) +
  geom_line(aes(y = vals,
                color = group,     
                linetype = group),
            lwd  = 1.5)  +
  theme_bw()

^{Created on 2021-04-19 by the reprex package (v1.0.0)}

Use group summary values as geom_ribbon in ggplot2?

1 Answers1

Linked