统计211

标题: R 学习笔记《一》~《十二》----R语言初学者指南 [打印本页]

作者: mxbj 时间: 2013-8-31 12:35
标题: R 学习笔记《一》~《十二》----R语言初学者指南
来源：http://www.dataguru.cn/article-3392-1.html

摘要: 本文汇总了开源中国社区的varlardoha博客里的R语言初学者指南《一》至《三》，里面有许多基础的总结，对初期学习的帮助还不错，有兴趣可看下。

《一》

第一章没啥可说的。

记录控制台几个命令：

？：       访问帮助恩见                         ?boxplot

# :       添加注释

boxplot: 生成盒形图

log:       自然对数                            log(2)

log10:       以10对底的对数                   log10(2)

library:    载入程序包                         library(MASS)

setwd:    设置工作目录                      setwd("c:/r/wd");注意斜杠是linux风格

q          关闭 R                            q()

citation 提供对R的引用                   citation()

以上操作直接打开R软件即可执行。

点击下载R for windows

点击下载Tinn-R编辑R代码

点击下载《R语言初学者指南》的源代码以及相关试验数据

《二》

1 录入少量数据。

打开Tinn R输入然后粘贴到R控制台

Wing1 <- 59
Wing2 <- 55
Wing3 <- 53.5
Wing4 <- 55
Wing5 <- 52.5
其实就是赋值运算，左边表示变量，<-符号也可以用=替换。

查看输入的值直接在R控制输入变量名然后回车即可

> Wing1
[1] 59
>
2 利用变量进行计算并赋值给新的变量

SQ.Wing1 <- sqrt(Wing1)
Mul.W1 <- 2*Wing1
Sum.12 <- Wing1+Wing2
SUM12345 <- Wing1+Wing2+Wing3+Wing4+Wing5
Av <- (Wing1 + Wing2+Wing3+Wing4+Wing5)/5

.是变量的一部分！

在R控制台输入Av

> Av
[1] 55
>

3  应用c函数连接数据

Wingcrd <- c(59,55,53.5,55,52.5,57.5,53,55)

> Wingcrd
[1] 59.0 55.0 53.5 55.0 52.5 57.5 53.0 55.0
>

取第一个元素

> Wingcrd[1]
[1] 59

输出前五个值
> Wingcrd[1:5]
[1] 59.0 55.0 53.5 55.0 52.5

删除第二个值，并显示剩余的值
> Wingcrd[-2]
[1] 59.0 53.5 55.0 52.5 57.5 53.0 55.0
>

4 通过R内置函数操作(sum,mean,max,min,median,var,sd)

在控制台输入

S.win <- sum(Wingcrd)
S.win

控制台的输出：

>
> S.win <- sum(Wingcrd)
> S.win
[1] 440.5
>
  5 输入其他数据

Tarsus <- c(22.3,19.7,20.8,20.3,20.8,21.5,20.6,21.5)
Head <- c(31.2,30.4,30.6,30.3,30.3,30.8,32.5,NA)
Wt <- c(9.5,13.8,14.8,15.2,15.5,15.6,15.6,15.7)

变量名一般大写字母开头

由于Head中有个数据是NA，因此调用内置函数时会出错

> sum(Head)
[1] NA
>

> sum(Head,na.rm = TRUE)
[1] 216.1
>
但是这样就不会出现NA了。原因：
向量中如果有一个确实值的话，默认选项na.rm=FALSE会导致R函数sum返回NA(rm表示移出)

采用sum(Head,na.rm = TRUE)即可计算出和。

na.rm=TRUE也可以应用到mean,min,max,median等函数中，一般可以用na.T替换na.TRUE选项。

但是不推荐。另外有些内置函数的选项可是是na.action而不是na.rm

《三》

1 使用 c,cbind,rbind结合变量

之前获得四列数据 Wingcrd,Tarsus,Head,Wt

每个列包含8个数据，可以通过c继续连接变量

> BirdData <- c(Wingcrd,Tarsus,Head,Wt)
> BirdData
  [1] 59.0 55.0 53.5 55.0 52.5 57.5 53.0 55.0 22.3 19.7 20.8 20.3 20.8 21.5
[15] 20.6 21.5 31.2 30.4 30.6 30.3 30.3 30.8 32.5 NA  9.5 13.8 14.8 15.2
[29] 15.5 15.6 15.6 15.7

BirdData是长度为32的单个向量，符号[1],[15],[29]不需要考虑不同电脑有不同的现实。这只是单个向量，R并没有区分这些值都属于哪一个变量。通过：

Id <- c(1,1,1,1,1,1,1,1,2,2,2,2,2,2,2,2,3,3,3,3,3,3,3,3,4,4,4,4,4,4,4,4)
Id <- rep(c(1,2,3,4),each = 8)
Id <- rep(1:4,each = 8)
这三个表达式效果是一样的

> Id <- rep(1:4,each = 8)
> Id
  [1] 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 4 4 4 4 4 4 4 4
>

a <- seq(from =1 ,to = 4,by = 1)
rep(a,each = 8)
效果：

> a <- seq(from =1 ,to = 4,by = 1)
> rep(a,each = 8)
  [1] 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 4 4 4 4 4 4 4 4
>

VarNames <- c("Wingcrd","Tarsus","Head","Wt")
Id2 <- rep(VarNames,each = 8)

> VarNames <- c("Wingcrd","Tarsus","Head","Wt")
> Id2 <- rep(VarNames,each = 8)
> Id2
  [1] "Wingcrd" "Wingcrd" "Wingcrd" "Wingcrd" "Wingcrd" "Wingcrd" "Wingcrd"
  [8] "Wingcrd" "Tarsus"  "Tarsus"  "Tarsus"  "Tarsus"  "Tarsus"  "Tarsus"
[15] "Tarsus"  "Tarsus"  "Head" "Head" "Head" "Head" "Head"
[22] "Head" "Head" "Head" "Wt"    "Wt"    "Wt"    "Wt"
[29] "Wt"    "Wt"    "Wt"    "Wt"
>

req(VarNames,8):

> rep(VarNames,8)
  [1] "Wingcrd" "Tarsus"  "Head" "Wt"    "Wingcrd" "Tarsus"  "Head"
  [8] "Wt"    "Wingcrd" "Tarsus"  "Head" "Wt"    "Wingcrd" "Tarsus"
[15] "Head" "Wt"    "Wingcrd" "Tarsus"  "Head" "Wt"    "Wingcrd"
[22] "Tarsus"  "Head" "Wt"    "Wingcrd" "Tarsus"  "Head" "Wt"
[29] "Wingcrd" "Tarsus"  "Head" "Wt"
>

cbind函数将结合的变量以列的形式输出

>
> Z <- cbind(Wingcrd,Tarsus,Head,Wt)
> Z
   Wingcrd Tarsus Head Wt
[1,] 59.0 22.3 31.2  9.5
[2,] 55.0 19.7 30.4 13.8
[3,] 53.5 20.8 30.6 14.8
[4,] 55.0 20.3 30.3 15.2
[5,] 52.5 20.8 30.3 15.5
[6,] 57.5 21.5 30.8 15.6
[7,] 53.0 20.6 32.5 15.6
[8,] 55.0 21.5 NA 15.7
>

访问Z的第一列

> Z[,1]
[1] 59.0 55.0 53.5 55.0 52.5 57.5 53.0 55.0
> Z[1:8,1]
[1] 59.0 55.0 53.5 55.0 52.5 57.5 53.0 55.0

访问Z的第一行

> Z[1,]
Wingcrd  Tarsus Head    Wt
59.0 22.3 31.2    9.5
> Z[1,1:4]
Wingcrd  Tarsus Head    Wt
59.0 22.3 31.2    9.5
>

同样可以使用的访问方式：

Z[1,]
Z[1,1:4]
Z[1,1]
Z[,2:3]
X <- Z[4,4]
Y <- Z[,4]
W <- Z[,3]得到第三列的数据
D <- Z[,c(1,3,4)]得到1,3,4列的所有数据
E <- Z[,c(-1,-3)] 负号表示排除第一第三列

显示Z的维数

dim(Z)

> dim(Z)
[1] 8 4
>

只查看Z行数

> Nrows <- dim(Z)[1]
> Nrows
[1] 8
>

rbind与cbind函数类似，只不过前者以行的形式表示数据，后者以列的形式表示数据

> Z2 <- rbind(Wingcrd,Tarsus,Head,Wt)
> Z2
      [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8]
Wingcrd 59.0 55.0 53.5 55.0 52.5 57.5 53.0 55.0
Tarsus  22.3 19.7 20.8 20.3 20.8 21.5 20.6 21.5
Head 31.2 30.4 30.6 30.3 30.3 30.8 32.5 NA
Wt    9.5 13.8 14.8 15.2 15.5 15.6 15.6 15.7
>

使用vector表示数据

> W <- vector(length = 8)
> W[1] <- 59
> W[2] <- 55
> W[3] <- 53.5
> W[4] <- 55
> W[5] <- 52.5
> W[6] <- 57.5
> W[7] <- 53
> W[8] <- 55
> W
[1] 59.0 55.0 53.5 55.0 52.5 57.5 53.0 55.0
>

说明：

在输入 W <- vector(length = 8)之后直接再输入W会得到一个FALSE的响应值。

必须在所有元素的值都输入之后才可以输入W查看数据

可以通过W[1],W[1:4],W[2:6],W[-2],W[c(1,2,3)]等查看数据

通过矩阵结合数据

> Dmat <- matrix(nrow = 8,ncol = 4)
> Dmat
   [,1] [,2] [,3] [,4]
[1,] NA NA NA NA
[2,] NA NA NA NA
[3,] NA NA NA NA
[4,] NA NA NA NA
[5,] NA NA NA NA
[6,] NA NA NA NA
[7,] NA NA NA NA
[8,] NA NA NA NA
>

初始化矩阵

> Dmat[,1] <- c(59,55,53.5,55,52.5,57.5,53,55)
> Dmat[,2] <- c(22.3,19.7,20.8,20.3,20.8,21.5,20.6,21.5)
> Dmat[,3] <- c(31.2,30.4,30.6,30.3,30.3,30.8,32.5,NA)
> Dmat[,4] <- c(9.5,13.8,14.8,15.2,15.5,15.6,15.6,15.7)
> Dmat
   [,1] [,2] [,3] [,4]
[1,] 59.0 22.3 31.2  9.5
[2,] 55.0 19.7 30.4 13.8
[3,] 53.5 20.8 30.6 14.8
[4,] 55.0 20.3 30.3 15.2
[5,] 52.5 20.8 30.3 15.5
[6,] 57.5 21.5 30.8 15.6
[7,] 53.0 20.6 32.5 15.6
[8,] 55.0 21.5 NA 15.7
>

使用colnames函数给矩阵Dmat的列添加名称

> colnames(Dmat) <- c("Wingcrd","Tarsus","Head","Wt")
> Dmat
   Wingcrd Tarsus Head Wt
[1,] 59.0 22.3 31.2  9.5
[2,] 55.0 19.7 30.4 13.8
[3,] 53.5 20.8 30.6 14.8
[4,] 55.0 20.3 30.3 15.2
[5,] 52.5 20.8 30.3 15.5
[6,] 57.5 21.5 30.8 15.6
[7,] 53.0 20.6 32.5 15.6
[8,] 55.0 21.5 NA 15.7
>

当数据按照变量进行了分类，则可以：

> Dmat2 <- as.matrix(cbind(Wingcrd,Tarsus,Head,Wt))
> Dmat2
   Wingcrd Tarsus Head Wt
[1,] 59.0 22.3 31.2  9.5
[2,] 55.0 19.7 30.4 13.8
[3,] 53.5 20.8 30.6 14.8
[4,] 55.0 20.3 30.3 15.2
[5,] 52.5 20.8 30.3 15.5
[6,] 57.5 21.5 30.8 15.6
[7,] 53.0 20.6 32.5 15.6
[8,] 55.0 21.5 NA 15.7
>

使用data.frame函数结合数据

使用数据框结合具有相同长度的变量，而数据框的每一行就包含有统一样本的不同观察值。

eg：

> Dfrm <- data.frame(WC = Wingcrd,TS = Tarsus,HD=Head,W=Wt)
> Dfrm
   WC TS HD W
1 59.0 22.3 31.2  9.5
2 55.0 19.7 30.4 13.8
3 53.5 20.8 30.6 14.8
4 55.0 20.3 30.3 15.2
5 52.5 20.8 30.3 15.5
6 57.5 21.5 30.8 15.6
7 53.0 20.6 32.5 15.6
8 55.0 21.5 NA 15.7
>

数据框的优点：可以在不改变原始数据的基础上改变数据。eg：

> Dfrm2 <- data.frame(WC = Wingcrd,TS = Tarsus,HD=Head,Wsq=sqrt(Wt))
> Dfrm2
   WC TS HD    Wsq
1 59.0 22.3 31.2 3.082207
2 55.0 19.7 30.4 3.714835
3 53.5 20.8 30.6 3.847077
4 55.0 20.3 30.3 3.898718
5 52.5 20.8 30.3 3.937004
6 57.5 21.5 30.8 3.949684
7 53.0 20.6 32.5 3.949684
8 55.0 21.5 NA 3.962323

Wt和W是不同的实体，验证:

> rm(Wt)
> Wt
错误: 找不到对象'Wt'
> Dfrm$W
[1]  9.5 13.8 14.8 15.2 15.5 15.6 15.6 15.7
>

数据框通常的用法:

向R中输入数据后对数据做些改变(移出极端值，应用变化，增加分类变量等等)，再将数据存入数据框中以备后续分析。

使用list结合数据

以list结合数据，list中的每一个数据既可以是向量，也可以是单个的数据等。其中向量的维数可能一样也可能不一样。

eg

> X1 <- c(1,2,3)
> X2 <- c("a","b","c","d")
> X3 <- 3
> X4 <- matrix(nrow = 2 , ncol = 2)
> X4[,1] <- c(1,2)
> X4[,2] <- c(3,4)
> Y <- list(LX1=X1,LX2=X2,LX3=X3,LX4=X4)
> Y
$LX1
[1] 1 2 3

$LX2
[1] "a" "b" "c" "d"

$LX3
[1] 3

$LX4
   [,1] [,2]
[1,] 1 3
[2,] 2 4

>

list的重要性：

线性回归，广义线性回归，t-检验等的结果一般都保存在list中

> M <- lm(WC ~ W,data = Dfrm)
> M

Call:
lm(formula = WC ~ W, data = Dfrm)

Coefficients:
(Intercept)          W
   65.5315    -0.7239
纤细分析的结果存储在：
> names(M)
  [1] "coefficients"  "residuals"    "effects"    "rank"
  [5] "fitted.values" "assign"       "qr"          "df.residual"
  [9] "xlevels"    "call"       "terms"       "model"
>

可以通过以下方式访问具体的值

> M$coefficients
(Intercept)          W
  65.5315140  -0.7238731
> M$residuals
      1       2       3       4       5       6       7
  0.3452800 -0.5420659 -1.3181928  0.4713564 -1.8114817  3.2609056 -1.2390944
      8
  0.8332929
> M$effects
  (Intercept)          W
-155.7402686 4.0250694 -1.2416235 0.5887546 -1.6634618

3.4191327 -1.0808673 1.0017273

>

综合：

> AllData <- list(BirdData = BirdData,Id = Id2,Z = Z,VarNames = VarNames)
> AllData
$BirdData
  [1] 59.0 55.0 53.5 55.0 52.5 57.5 53.0 55.0 22.3 19.7 20.8 20.3 20.8 21.5
[15] 20.6 21.5 31.2 30.4 30.6 30.3 30.3 30.8 32.5 NA  9.5 13.8 14.8 15.2
[29] 15.5 15.6 15.6 15.7

$Id
  [1] "Wingcrd" "Wingcrd" "Wingcrd" "Wingcrd" "Wingcrd" "Wingcrd" "Wingcrd"
  [8] "Wingcrd" "Tarsus"  "Tarsus"  "Tarsus"  "Tarsus"  "Tarsus"  "Tarsus"
[15] "Tarsus"  "Tarsus"  "Head" "Head" "Head" "Head" "Head"
[22] "Head" "Head" "Head" "Wt"    "Wt"    "Wt"    "Wt"
[29] "Wt"    "Wt"    "Wt"    "Wt"

$Z
   Wingcrd Tarsus Head Wt
[1,] 59.0 22.3 31.2  9.5
[2,] 55.0 19.7 30.4 13.8
[3,] 53.5 20.8 30.6 14.8
[4,] 55.0 20.3 30.3 15.2
[5,] 52.5 20.8 30.3 15.5
[6,] 57.5 21.5 30.8 15.6
[7,] 53.0 20.6 32.5 15.6
[8,] 55.0 21.5 NA 15.7

$VarNames
[1] "Wingcrd" "Tarsus"  "Head" "Wt"

分别取其中的某一个元素

> AllData$BirdData
  [1] 59.0 55.0 53.5 55.0 52.5 57.5 53.0 55.0 22.3 19.7 20.8 20.3 20.8 21.5
[15] 20.6 21.5 31.2 30.4 30.6 30.3 30.3 30.8 32.5 NA  9.5 13.8 14.8 15.2
[29] 15.5 15.6 15.6 15.7
> AllData$Id
  [1] "Wingcrd" "Wingcrd" "Wingcrd" "Wingcrd" "Wingcrd" "Wingcrd" "Wingcrd"
  [8] "Wingcrd" "Tarsus"  "Tarsus"  "Tarsus"  "Tarsus"  "Tarsus"  "Tarsus"
[15] "Tarsus"  "Tarsus"  "Head" "Head" "Head" "Head" "Head"
[22] "Head" "Head" "Head" "Wt"    "Wt"    "Wt"    "Wt"
[29] "Wt"    "Wt"    "Wt"    "Wt"
> AllData$Z
   Wingcrd Tarsus Head Wt
[1,] 59.0 22.3 31.2  9.5
[2,] 55.0 19.7 30.4 13.8
[3,] 53.5 20.8 30.6 14.8
[4,] 55.0 20.3 30.3 15.2
[5,] 52.5 20.8 30.3 15.5
[6,] 57.5 21.5 30.8 15.6
[7,] 53.0 20.6 32.5 15.6
[8,] 55.0 21.5 NA 15.7
>

注意：在list中只能使用= 不能使用 <-

欢迎光临统计211 (http://www.tj211.com/)