2013-08-27 39 views
0

我有8個變量(X1,X2 ... X8)的數據幀列表離羣值在數據幀

我想使用獲得的箱線圖的離羣值:

boxplot(dataframe, plot=FALSE)$out 

我的期望輸出是讓數據框列出每個變量的異常值。具體如下:

variable outlier 
x1  outlier1 from x1 
x1  outlier2 from x1 
x1  outlier3 from x1 
x1  outlier4 from x1 
x2  outlier1 from x2 
x2  outlier2 from x2 
x2  outlier3 from x2 
. 
. 
. 
x8  outliern from x8 

感謝您的支持,

回答

1

這是你想要的嗎?

> testdata <- data.frame(x1=runif(1e3),x2=rnorm(1e3),x3=rnorm(1e3)) 
> temp <- boxplot(testdata,plot=F) 
> cbind(temp$group,temp$out) 
     [,1]  [,2] 
[1,] 2 2.765277 
[2,] 2 2.754730 
[3,] 2 -2.714811 
[4,] 2 3.257889 
[5,] 3 2.605549 
[6,] 3 -3.261950 
[7,] 3 -3.057532 
[8,] 3 2.820352 
[9,] 3 2.602933 
[10,] 3 2.580897 
[11,] 3 2.899350