R数据科学第五章
library(tidyverse)
变量的分布进行可视化
ggplot(data = diamonds) +
geom_bar(mapping = aes(x = cut))
图片.png
- 观测值可以使用dplyr::count()计算统计
diamonds %>%
count(cut)
图片.png
连续变量，用直方图
ggplot(data = diamonds) +
geom_histogram(mapping = aes(x = carat), binwidth = 0.5)
图片.png
-
当然也可以用dplry::count和ggplot::cut_width进行手动统计
diamonds %>% count(cut_width(carat, 0.5))
图片.png 根据自己的目的选择直方图的宽度，如果选择小于3克拉的钻石
smaller <- diamonds %>%
filter(carat < 3)
ggplot(data = smaller, mapping = aes(x = carat)) +
geom_histogram(binwidth = 0.1)
图片.png
- 如果使用叠加的条形图则用geom_freqpoly代替geom_histogram,但前者是折线图统计
ggplot(data = smaller, mapping = aes(x = carat, color = cut)) +
geom_histogram(binwidth = 0.1)
ggplot(data = smaller, mapping = aes(x = carat, color = cut)) +
geom_freqpoly(binwidth = 0.1)
ggplot(data = smaller, mapping = aes(x = carat)) +
geom_histogram(binwidth = 0.1)
ggplot(data = faithful, mapping = aes(x = eruptions)) +
geom_histogram(binwidth = 0.25)
图片.png
图片.png
5.3异常值
ggplot(diamonds) +
geom_histogram(mapping = aes(x = y), binwidth = 0.5)
图片.png
- coord_cartesian（）放到靠近0的数值
ggplot(diamonds) +
geom_histogram(mapping = aes(x = y), binwidth =0.5) +
coord_cartesian(ylim = c(0,50))
图片.png
- 可以看到有三个异常的数值,利用dplry中的filter将他们 找到
unusual <- diamonds %>%
filter(y < 3 | y > 20) %>%
arrange(y)
unusual
图片.png
ggplot(unusual) +
geom_histogram(mapping = aes(x = y), binwidth =0.5) +
coord_cartesian(ylim = c(0,50))
图片.png
钻石为0.99克拉和1克拉的数量，为什么出现这样的结果
m0.99 <- diamonds %>%
filter(carat == 0.99)
m0.99
m1 <- diamonds %>%
filter(carat == 1)
m1
图片.png
- 缺失值
diamonds2 <- diamonds %>%
filter(between(y, 3, 20)) %>%
arrange(y)
diamonds2
- 利用缺失值NA代替异常值,mutate()
diamonds3 <- diamonds %>%
mutate(y = ifelse(y < 3|y >20, NA, y))
diamonds3
- ggplot2中遵循无视缺失值的原则，忽略缺失值
ggplot(data = diamonds3, mapping = aes(x = x, y = y)) +
geom_point()
###Warning message:
Removed 9 rows containing missing values (geom_point).
- 警告忽略缺失值，可用na.rn = TURE,消除警告
ggplot(data = diamonds3, mapping = aes(x = x, y = y)) + geom_point(na.rm = TRUE)
图片.png
- 下边这个没看懂 原文是：弄清楚造成缺失值的观测和没有缺失值的观测间的区别的原因，例如：在nycflights13::flights 中，dep_time变量中的缺失值表示 航班取消了，因子，应该比较一下 已取消的航班和未取消航班的计划出发时间，利用is.na()函数创建一个新变量来完成这个操作
nycflights13::flights %>%
mutate(
cancelled = is.na(dep_time),
sched_hour = sched_dep_time %/% 100,
sched_min = sched_dep_time %% 100,
sched_dep_time = sched_hour + sched_min / 60
) %>%
ggplot(mapping = aes(sched_dep_time)) +
geom_freqpoly(
mapping = aes(color = cancelled),
binwidth = 1/4)
图片.png
相关变动是两个或者多个变量以相关的方式共同变化所表现出的趋势
- ** 分类变量和连续变量**
ggplot(data = diamonds, mapping = aes(x = price)) +
geom_freqpoly(mapping = aes(color = cut), binwidth = 500)
图片.png
- 注：应为数量差别很大和难看出差距
- 将纵坐标改成密度 density，相当于对计数进行了标准化，这样每个频率多边形下面的面积都是1
ggplot(data = diamonds, mapping = aes(x = price, y = ..density..)) +
geom_freqpoly(mapping = aes(color = cut), binwidth = 500)
图片.png
箱线图可以将分类变量进行可视化
- geom_boxplot函数查看切割质量和价格分布
ggplot(data = diamonds, mapping = aes(x = cut, y = price)) +
geom_boxplot(mapping = aes(color = cut)
)
图片.png
- 利用reorder()函数进行排列
ggplot(data = mpg, mapping = aes(x = class, y = hwy)) +
geom_boxplot()
ggplot(data = mpg, mapping = aes(x = reorder(class, hwy, FUN = median), y = hwy)) +
geom_boxplot()
图片.png
图片.png
- 利用coord_flip将函数图形旋转90度
ggplot(data = mpg, mapping = aes(x = reorder(class, hwy, FUN = median), y = hwy)) +
geom_boxplot() +
coord_flip()
图片.png
个人学习笔记，记录的不够详细，比较粗糙，勿喷。