> names(hr)
[1] "satisfaction" "evaluation" "project" "monthlyhour" "serviceyear" "accident" "left" "promotion" "dept" "salary" "SG"
变量选取的三种方法如下：
1.subset( )
eg：
> hr01<-subset(hr, select=c(left,salary) )
> names(hr01)
left salary
> hr02<-subset(hr, select=left：salary)
> names(hr02)
left promotion dept salary
> hr03<-subset(hr, select = - salary)
> names(hr03)
satisfaction evaluation project monthlyhour serviceyear accident left promotion dept SG
2.dplyr包——filter()
filter()函数并不能选取我们感兴趣的变量，只能按照条件筛选出符合要求的cases，
> hr04<-filter(hr, hr$evaluation>0.72 & hr$project>4 | hr$serviceyear>=4)
3.dplyr包——select()
> hr05<-select(hr,start_with("")) 筛选出名称中以xx开头的变量
> hr06<-select(hr,end_with("")) 筛选出名称中以xx结尾的变量
> hr07<-select(hr,contains("")) 筛选出名称中包含xx的变量
> hr08<-select(hr,-start_with("")) 剔除掉名称中以xx开头的变量
> hr_top1<-hr_top %>%select(satisfaction:promotion,SG)
顺带提一下dplyr包的这个管道函数“%>%”，管道函数可以将左边的对象传递给右边的函数。
转载于:https://www.cnblogs.com/yilingjian/p/7281169.html
