R语言学习（五）基本图形

条形图

条形图通过垂直的或者水平的条形展示了类别型变量的分布
barplot(height) height是一个矩阵或者向量
接下来的示例中我们用到的数据是包含在vcd包中的Arthitis数据集，是关于风湿性关节炎新疗法的研究结果
在首次运行时，先安装这个包

> install.packages("vcd")

其中Arthitis数据集的Improved变量记录了安慰剂或治疗药物对病人的治疗结果

> library(vcd)
载入需要的程辑包：grid
> Arthritis
   ID Treatment    Sex Age Improved
1  57   Treated   Male  27     Some
2  46   Treated   Male  29     None
3  77   Treated   Male  30     None
4  17   Treated   Male  32   Marked
5  36   Treated   Male  46   Marked
6  23   Treated   Male  58   Marked
7  75   Treated   Male  59     None
8  39   Treated   Male  59   Marked
9  33   Treated   Male  63     None
10 55   Treated   Male  63     None
11 30   Treated   Male  64     None
12  5   Treated   Male  64     Some
13 63   Treated   Male  69     None
14 83   Treated   Male  70   Marked
15 66   Treated Female  23     None
16 40   Treated Female  32     None
17  6   Treated Female  37     Some
18  7   Treated Female  41     None
19 72   Treated Female  41   Marked
20 37   Treated Female  48     None
21 82   Treated Female  48   Marked
22 53   Treated Female  55   Marked
23 79   Treated Female  55   Marked
24 26   Treated Female  56   Marked
25 28   Treated Female  57   Marked
26 60   Treated Female  57   Marked
27 22   Treated Female  57   Marked
28 27   Treated Female  58     None
29  2   Treated Female  59   Marked
30 59   Treated Female  59   Marked
31 62   Treated Female  60   Marked
32 84   Treated Female  61   Marked
33 64   Treated Female  62     Some
34 34   Treated Female  62   Marked
35 58   Treated Female  66   Marked
36 13   Treated Female  67   Marked
37 61   Treated Female  68     Some
38 65   Treated Female  68   Marked
39 11   Treated Female  69     None
40 56   Treated Female  69     Some
41 43   Treated Female  70     Some
42  9   Placebo   Male  37     None
43 14   Placebo   Male  44     None
44 73   Placebo   Male  50     None
45 74   Placebo   Male  51     None
46 25   Placebo   Male  52     None
47 18   Placebo   Male  53     None
48 21   Placebo   Male  59     None
49 52   Placebo   Male  59     None
50 45   Placebo   Male  62     None
51 41   Placebo   Male  62     None
52  8   Placebo   Male  63   Marked
53 80   Placebo Female  23     None
54 12   Placebo Female  30     None
55 29   Placebo Female  30     None
56 50   Placebo Female  31     Some
57 38   Placebo Female  32     None
58 35   Placebo Female  33   Marked
59 51   Placebo Female  37     None
60 54   Placebo Female  44     None
61 76   Placebo Female  45     None
62 16   Placebo Female  46     None
63 69   Placebo Female  48     None
64 31   Placebo Female  49     None
65 20   Placebo Female  51     None
66 68   Placebo Female  53     None
67 81   Placebo Female  54     None
68  4   Placebo Female  54     None
69 78   Placebo Female  54   Marked
70 70   Placebo Female  55   Marked
71 49   Placebo Female  57     None
72 10   Placebo Female  57     Some
73 47   Placebo Female  58     Some
74 44   Placebo Female  59     Some
75 24   Placebo Female  59   Marked
76 48   Placebo Female  61     None
77 19   Placebo Female  63     Some
78  3   Placebo Female  64     None
79 67   Placebo Female  65   Marked
80 32   Placebo Female  66     None
81 42   Placebo Female  66     None
82 15   Placebo Female  66     Some
83 71   Placebo Female  68     Some
84  1   Placebo Female  74   Marked
> counts<-table(Arthritis$Improved)
> counts

  None   Some Marked 
    42     14     28 

先来绘制简单的条形图此时counts是一个向量

> barplot(counts,main = "Simple Bar Plot",xlab = "Improvement",ylab = "Frequency")

image.png

> barplot(counts,main = "Simple Bar Plot",xlab = "Improvement",ylab = "Frequency",horiz = TRUE) #水平条形图，默认为垂直，添加horiz=TRUE可以改为水平

image.png

如果counts是一个矩阵而不是向量，那么绘图结果将是堆砌条形图或分组条形图

> counts<-table(Arthritis$Improved,Arthritis$Treatment)
> counts
        
         Placebo Treated
  None        29      13
  Some         7       7
  Marked       7      21

> barplot(counts,main = "Stacked Bar Plot",xlab = "Treatment",ylab = "Frequency",col = c("yellow","red","green"),legend=rownames(counts))

image.png

分组条形图（较为常用）

> barplot(counts,main = "Stacked Bar Plot",xlab = "Treatment",ylab = "Frequency",col = c("yellow","red","green"),legend=rownames(counts),beside = TRUE) #beside=TRUE 选定为分组条形图，默认为堆砌条形图

image.png

饼图

还有一类图可以用来表示类别型变量，但是相对于饼图，多数统计学家更推荐使用条形图或者点图，因为相对于面积，人们对于长度的判断往往更精确，所以这里就先不学习饼图了

直方图

直方图通过在X轴上将值域分割为一定数量的组，在Y轴上显示相应值的频数，展示了连续型变量的分布
hist(x),x是一个由数值组成的数值向量
在下面的示例中，我们用到的数据集是R语言的内置数据集mtcars

> par(mfrow=c(1,2))
> hist(mtcars$mpg)
> hist(mtcars$mpg,breaks = 12,col = "red",xlab = "Miles Per Gallon",main = "Colored histogram with 12 bins")

image.png

核密度图

核密度图是估计随机变量概率密度的一种非参数方法
plot(density(x))

> plot(density(mtcars$mpg))

image.png

箱线图

箱线图通过绘制连续型变量的五数，即最小值，下四分位数，中位数，上四分位数，最大值来描述连续型变量的分布
boxplot(x)

> boxplot(mtcars$mpg,main="Box plot",ylab="Miles Per Gallon")

image.png

从下到上五条线依次为最小值，下四分位数，中位数，上四分位数，最大值
默认情况下，两条须的延申不会超过盒型各端加1.5倍四分位的距离，此外的离群点将以点来表示

并列箱线图

箱线图可以展示单个变量或分组变量
boxplot(formula,data=dataframe)
formual是一个公式，dataframe则是提供的数据框
一个示例公式Y_{A，表示类别型变量A的每个值生成数值型变量Y的箱线图，就是Y作为Y轴，A作为X轴，Y}A*B，由A和B所有水平的两两组合生成Y

> boxplot(mpg~cyl,data = mtcars,main="Car Mileage Data",xlab="Number of cylinders",ylab="Miles per gallon")

image.png

从图可以看书随着发动机缸数的增多，每加仑油行驶的英里数下降，耗油增加

点图

点图提供了一种在简单水平上绘制大量有标签值的方法
dotchart(x,labels=)
x是一个数值向量，labels是一个由每个点的标签组成的向量
可以通过添加参数groups来选定一个因子，用以指定x中的分组方式
gcolor参数可以控制不同组的标签的颜色
cex控制标签大小

> dotchart(mtcars$mpg,labels = rownames(mtcars),cex = .7,main = "Gas Mileage for car models",xlab = "Miles per gallon")

image.png

通常经过排序，分组后的点图更具有观察效果，下面我们根据mpg数值从小到大进行排序并根据汽车发动机缸数（cyl）的个数来分组

> x<-mtcars[order(mtcars$mpg),]#对mtcars按照mpg大小排序，赋给x
> x$cyl<-factor(x$cyl)#将cyl从字符向量转化为因子向量
#在x中新建一个color的字符向量
> x$color[x$cyl==4]<-"red"
> x$color[x$cyl==6]<-"blue"
> x$color[x$cyl==8]<-"darkgreen"
> dotchart(x$mpg,labels = rownames(x),cex = .7,groups = x$cyl,gcolor = "black",color = x$color,pch = 19,main = "Gas Mileage for Car Models",xlab = "Mile Per Gallon")

image.png

这个图相比上一个图结果更加明显，缸数少的汽车每加仑行驶的英里数增加，耗油量减少