我在RStudio中使用R
,并且具有以下数据框。
df1 <- data.frame(
comp = c("A", "B", "C", "D", "E", "F"),
Q2_2018 = c(27, 10, 6, 4, 3, 2),
Q2_2019 = c(31, 12, 8, 6, 5, 4))
我想根据上面的数据创建一个图表,如下所示(不包括Amazon徽标)。
我主要停留在绘制具有%变化的圆圈。
到目前为止,
library(ggplot2)
library(reshape2)
library(magrittr)
melt(df1, id.vars = "comp") %>%
ggplot(aes(x= comp, y=value, fill=variable)) + geom_bar(stat = "identity", position = "dodge")
可以用ggplot2
完成吗?
答案 0 :(得分:1)
大部分方式:
library(tidyverse)
df1 %>%
gather(year, val, -comp) %>%
group_by(comp) %>%
mutate(change = val / lag(val) - 1) %>%
mutate(change_lab = if_else(!is.na(change),
scales::percent(change,
accuracy = 1,
prefix = if_else(change > 0, "+", "-")),
NA_character_)) %>%
ungroup() %>%
ggplot(aes(comp, val, fill = year, label = val)) +
geom_col(position = position_dodge()) +
geom_text(position = position_dodge(width = 1), vjust = -0.5) +
geom_point(aes(comp, val + 5, size = change), color = "lightgreen") +
geom_text(aes(comp, val+5, label = change_lab)) +
scale_size_area(max_size = 30) +
guides(size= F) +
theme_classic()