CODEWITHSUNDEEP
rggplot2labelaxis
Ant
Ant

Reputation: 343

Mixed format axis labels in ggplot chart

Is it possible to have a ggplot axis with date-formatted labels and text-formatted labels?

The following code produces a chart with a date formatted x axis that runs from Jan-18 to Jun-19. 

#create plot with line and points.
#colour of points based on colour column of meltdf2
ggplot(data = meltdf2, aes(x = Month, y = variable, group = variable)) +
  geom_line(linetype = "dashed", colour = "grey") +
  geom_point(aes(colour = meltdf2$colour, size = 3)) +
  geom_point(data = melt_change, aes(shape = value, size = 3)) +
  scale_colour_manual(values = pal, limits = names(pal), guide_legend("Risk Level")) +
  scale_shape_manual(values = sh_pal, limits = names(sh_pal), guide_legend("Latest Change")) +
  scale_x_date(date_breaks = "1 month" , date_labels = "%b-%y") +
  theme(axis.title.y=element_blank(), axis.text.y=element_blank(), axis.ticks.y=element_blank(),
        axis.line.x = element_line(colour = "darkgrey"), axis.title.x=element_blank(),
        panel.background = element_rect(fill = "white"),
        panel.grid.major = element_blank(), panel.grid.minor = element_blank(),
        legend.position = c(0.2,0.70),
        legend.key.size = unit(1.5,"line")) + 
  guides(colour = guide_legend(override.aes = list(size = 3)), shape = guide_legend(override.aes = list(size = 5))) +
  scale_size(guide = "none")

Chart

However, I would like the Jun-19 date label to be replaced by a text label that reads "Change". Is there a way to do this without specifying all the labels individually?

I have tried to produce a vector that can be used as a labels argument in the scale_x_date function using the following code:

x_labels <- (as.Date(unique(meltdf2$Month)))
x_labels <- sort(x_labels)
x_labels <- c(x_labels,"Change")

But the vector does not seem to sort the dates correctly and the text "Change" value is replaced by an NA.

Upvotes: 1

Views: 287

Answers (1)

bbiasi
bbiasi

Reputation: 1599

As the example is not fully reproducible, I have created a response with fictitious data.

Note that I had to create a new column for the period (Period) under analysis, so that it is a factor and where I added the name "Change".

Later I reordered the levels using forcats, function fct_reorder, based on the dates (Dates) of interest.

library(dplyr)
library(lubridate)
library(ggplot2)
library(forcats)
set.seed(1)
Period_x <- data.frame(Period = seq(lubridate::mdy('jan-01-2018'),
                                    lubridate::mdy('may-01-2019'), 
                                    by = 'month'))

Period_x <- Period_x %>% 
  dplyr::mutate(Period = format(Period, '%b-%y'))
Period_x <- rbind(Period_x, "Change")

Period2 <- seq(lubridate::mdy('jan-01-2018'),
               lubridate::mdy('jun-01-2019'), 
               by = 'month')

Period_x <- Period_x %>% 
  dplyr::mutate(Dates = Period2,
                y = rnorm(18), 
                Period = as.factor(Period),
                Period = forcats::fct_reorder(Period, 
                                              Dates))

ggplot2::ggplot(Period_x) +
  geom_point(aes(x = Period, y = y))

enter image description here

Upvotes: 1

Related Questions