使用包中的diamonds数据集ggplot2,我可以生成以下图表.
library(ggplot2)
library(dplyr)
diamond.summary <-
diamonds %>%
mutate(carat = ifelse(runif(nrow(.)) < 0.05, NA_real_, carat)) %>%
group_by(carat_quintile = ntile(carat, 5)) %>%
summarise(avg_price = mean(price))
diamond.summary %>%
filter(!is.na(carat_quintile)) %>%
ggplot(aes(carat_quintile, avg_price)) +
geom_bar(stat = "identity",
color = "black",
width = 1) +
scale_x_continuous("Carat percentile",
breaks = 1:6 - 0.5,
labels = seq(0,100, by = 20)) +
scale_y_continuous(expand = c(0,0),
limits = c(0, 1.1* max(diamond.summary$avg_price)))
Run Code Online (Sandbox Code Playgroud)
到目前为止,这么容易.但是,我还想在图表旁边显示缺失条目的平均价格.类似于以下内容:

diamond.summary %>%
mutate(Facet = is.na(carat_quintile),
carat_quintile_noNA = ifelse(Facet, "Unknown", carat_quintile)) %>%
ggplot(aes(x = carat_quintile_noNA, y = avg_price, fill = Facet)) +
geom_bar(stat = "identity") +
facet_grid(~Facet, scales = "free_x", space = "free_x") +
scale_x_discrete(breaks = (0:6) - 0.5)
Run Code Online (Sandbox Code Playgroud)
但是,当我尝试使用相同的技巧时scale_x_continuous,我得到了错误Discrete value supplied to continuous scale.scale_x_discrete(breaks = c(0:6 + 0.5))例如,当我尝试使用时,轴刻度和标签消失.
我的问题是,如何将第一个面板中的刻度线放在上面的同一个刻面图表中,如同在这篇文章的第一个图表中一样?关于图表设计的建议可能是一个可接受的解决方案,但我不认为这样的所有问题都可以通过重新设计来解决.
诀窍是将您的因子转换为数字,为未知数量指定幻数.(ggplot2不会绘制具有真NA值的条形图.)然后使用scale_x_continuous
diamond.summary %>%
mutate(Facet = is.na(carat_quintile),
carat_quintile_noNA = ifelse(Facet, "Unknown", carat_quintile),
##
## 99 is a magic number. For our plot, it just has
## to be larger than 5. The value 6 would be a natural
## choice, but this means that the x tick marks would
## overflow ino the 'unknown' facet. You could choose
## choose 7 to avoid this, but any large number works.
## I used 99 to make it clear that it's magic.
numeric = ifelse(Facet, 99, carat_quintile)) %>%
ggplot(aes(x = numeric, y = avg_price, fill = Facet)) +
geom_bar(stat = "identity", width = 1) +
facet_grid(~Facet, scales = "free_x", space = "free_x") +
scale_x_continuous(breaks = c(0:5 + 0.5, 99),
labels = c(paste0(c(0:5) * 20, "%"), "Unknown"))
Run Code Online (Sandbox Code Playgroud)