数据的筛选

要提取相应内容的数据，最为常用的就是提取相应元素，比如提取某个元素，提取某一行，某一列。我们通过下面下面的例子来学习：

data<-data.frame(a=sample(1:10),b=rep(c("a","b"),each=5),cdf=rnorm(10))data

##     a b     cdf
## 1   1 a  0.5755
## 2  10 a  0.8087
## 3   2 a  0.9810
## 4   7 a -0.4635
## 5   4 a  0.5094

#提取相应元素data[2,1]

## [1] 10

data[[1]][[2]]

## [1] 10

data[[c(1,2)]]

## [1] 10

data$a[2]

## [1] 10

#提取某一列data[[3]]

##  [1]  0.5755  0.8087  0.9810 -0.4635  0.5094  1.0514 -1.5338  1.0047
##  [9]  1.0004 -1.3566

data$cdf

##  [1]  0.5755  0.8087  0.9810 -0.4635  0.5094  1.0514 -1.5338  1.0047
##  [9]  1.0004 -1.3566

data$c

##  [1]  0.5755  0.8087  0.9810 -0.4635  0.5094  1.0514 -1.5338  1.0047
##  [9]  1.0004 -1.3566

data[["c"]]

## NULL

data[["c", exact = FALSE]]

##  [1]  0.5755  0.8087  0.9810 -0.4635  0.5094  1.0514 -1.5338  1.0047
##  [9]  1.0004 -1.3566

数据的筛选还有一个最为常用的的就是移除缺失值：

data<-data.frame(a=c(sample(1:5),NA,NA,sample(6:10)),b=c(rep(c("a","b"),each=5),NA,NA),cdf=rnorm(12))data

##     a    b       cdf
## 1   5    a -0.276400
## 2   1    a -1.861240

good <- complete.cases(data)data[good, ]

##    a b     cdf
## 1  5 a -0.2764
## 2  1 a -1.8612
## 3  3 a -2.0280

bad <- as.data.frame(is.na(data))data[!(bad$a|bad$b|bad$c),]

##    a b     cdf
## 1  5 a -0.2764
## 2  1 a -1.8612

数据筛选有时是为了获得符合条件的数据：

X <- data.frame("var1"=sample(1:5),"var2"=sample(6:10),"var3"=sample(11:15))X <- X[sample(1:5),]; X$var2[c(1,3)] = NAX

##   var1 var2 var3
## 2    5   NA   13
## 5    3    6   15
## 1    2   NA   12
## 3    1    8   11
## 4    4    9   14

X[(X$var1 <= 3 & X$var3 > 11),]

##   var1 var2 var3
## 5    3    6   15
## 1    2   NA   12

subset(X,(X$var1 <= 3 & X$var3 > 11))

##   var1 var2 var3
## 5    3    6   15
## 1    2   NA   12

X[(X$var1 <= 3 | X$var3 > 15),]

##   var1 var2 var3
## 5    3    6   15
## 1    2   NA   12
## 3    1    8   11

X[which(X$var1 <= 3 | X$var3 > 15),]

##   var1 var2 var3
## 5    3    6   15
## 1    2   NA   12
## 3    1    8   11

对于取子集的函数subset，在帮助文档中有一段warning是值得我们注意的：“This is a convenience function intended for use interactively. For programming it is better to use the standard subsetting functions like [, and in particular the non-standard evaluation of argument subset can have unanticipated consequences."

分享到： QQ空间 新浪微博 腾讯微博 人人网 飞信 百度搜藏