使用ggplot散布点 [英] Using ggplot for scattering dots
问题描述
我得到了以下数据集:
Name Year-Month Value
A 2002-01 -3.45
A 2003-02 2.87
A 2004-05 1.78
A 2005-01 -9.54
B 2000-01 -1.45
B 2001-02 10.87
B 2002-01 5.78
C 2004-01 -6.45
C 2005-01 4.87
我想要做的是我想以特殊方式绘制值. 在x轴上应该有年份和月份，但是由于在2000-2008年之间有很多年份和月份的观测值，我只想写出一年中的第一个月和第六个月，而一年中的其他所有月份都是只是用一个标志标记.
What I want to do is I want to plot the values in special way. On the x-Axis there should be the years and months, but as there are many observations from years and months between 2000-2008 I only want to write out the first and six month of the year and all the other months of the year are just marked by a sign.
对于所有观察，我都希望将值分散成点或十字形，而不管它们来自哪个字母.
For all observations I want to scatter the values like a dot or a cross, not matter from which letter they are.
我画了一张照片.
这张照片只是一个片段.在实际图中，值必须精确地在年月所在的行上.
This picture is just a scatch. In the real plot, the values have to be excatly on the line where the year-month is.
是否有使用ggplot2或其他任何软件包执行此操作的简便方法?
Is there an easy way of doing this with ggplot2 or any other package?
推荐答案
要每6个月显示一次x轴标签，同时每月显示次要刻度线，我们需要一些技巧:每个月进行一次主要刻度线，但只显示每6个月贴标签.
To display the x-axis labels every 6 months while showing the minor ticks every month, we need a little trick: make major ticks for every month but only show labels every 6 months.
要使用scale_x_date，需要从Year-Month创建一个伪" Date列.在这里，我只是将每月01的第一天添加到现有的Year-Month列中.
To make use of scale_x_date, creating a "fake" Date column from Year-Month is needed. Here I just append the first day of the month 01 to the existing Year-Month column.
library(magrittr)
library(tidyverse)
df <- read.table(text = "Name Year-Month Value
A 2002-01 -3.45
A 2003-02 2.87
A 2004-05 1.78
A 2005-01 -9.54
B 2000-01 -1.45
B 2001-02 10.87
B 2002-01 5.78
C 2004-01 -6.45
C 2005-01 4.87",
header = TRUE)
# Create a Date column so that scale_x_date can be used
df %<>%
as.tibble() %>%
mutate(Date = as.Date(paste0(Year.Month, "-01")))
df
#> # A tibble: 9 x 4
#> Name Year.Month Value Date
#> <fct> <fct> <dbl> <date>
#> 1 A 2002-01 -3.45 2002-01-01
#> 2 A 2003-02 2.87 2003-02-01
#> 3 A 2004-05 1.78 2004-05-01
#> 4 A 2005-01 -9.54 2005-01-01
#> 5 B 2000-01 -1.45 2000-01-01
#> 6 B 2001-02 10.9 2001-02-01
#> 7 B 2002-01 5.78 2002-01-01
#> 8 C 2004-01 -6.45 2004-01-01
#> 9 C 2005-01 4.87 2005-01-01
# Auto x-axis break
ggplot(df, aes(x = Date, y = Value)) +
geom_point(pch = 4, size = 5) +
scale_x_date(expand = c(0.015, 0.015),
breaks = scales::pretty_breaks(), date_labels = "%Y-%b") +
theme_bw()
# Break every 6 months
ggplot(df, aes(x = Date, y = Value)) +
geom_point(pch = 4, size = 5) +
scale_x_date(expand = c(0.015, 0.015),
date_breaks = "6 months", date_labels = "%Y-%b") +
theme_bw()
# Color by Name, manually setup date range
ggplot(df, aes(x = Date, y = Value, color = Name)) +
geom_point(pch = 4, size = 5) +
scale_x_date(expand = c(0.015, 0.015),
breaks = seq(min(df$Date), max(df$Date), by = "6 months"),
date_minor_breaks = "1 month",
date_labels = "%Y-%b") +
theme_bw()
# Add minor tick
# Trick: make major ticks for every month but only show labels every 6 months
labels_month = format(seq(from = min(df$Date), to = max(df$Date), by = "1 months"),
"%Y-%b")
labels_month[rep(c(FALSE, TRUE), c(1, 4))] <- ""
labels_month
#> [1] "2000-Jan" "" "" "" "" "2000-Jun"
#> [7] "" "" "" "" "2000-Nov" ""
#> [13] "" "" "" "2001-Apr" "" ""
#> [19] "" "" "2001-Sep" "" "" ""
#> [25] "" "2002-Feb" "" "" "" ""
#> [31] "2002-Jul" "" "" "" "" "2002-Dec"
#> [37] "" "" "" "" "2003-May" ""
#> [43] "" "" "" "2003-Oct" "" ""
#> [49] "" "" "2004-Mar" "" "" ""
#> [55] "" "2004-Aug" "" "" "" ""
#> [61] "2005-Jan"
x_breaks = seq(min(df$Date), max(df$Date), by = "1 months")
ggplot(df, aes(x = Date, y = Value, color = Name)) +
geom_point(pch = 4, size = 5) +
scale_x_date(expand = c(0.015, 0.015),
labels = labels_month,
breaks = x_breaks) +
theme_classic() +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5))
由 reprex软件包(v0.2.0)创建于2018-06-05.
Created on 2018-06-05 by the reprex package (v0.2.0).
这篇关于使用ggplot散布点的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！
