與isplit

Question

塊引導我有以下示例數據集

Sites<-c("A","A","B","B","B","C","C","C") 
TR<-c(0.1,0.4,0.4,0.3,0.3,0.6,0.5,0.5) 
df<-data.frame(Sites,TR) 

我想由變量到子集的行「站點」，然後引導每個站點捕集率（TR）。我曾嘗試使用下面的代碼

require(iterators) 
sites<-isplit(df,df$Sites) 
nextElem(sites) 
require(foreach) 
mT<-foreach(Sites=sites)%do% 

{ 
meanTraps <- vector() 
for(i in 1:boots){ 
    meanTraps[i] <- mean(sample(sites$value$TR, nrow(sites$value), replace = TRUE)) 
} 
} 

試圖但這恰恰返回

list() 

理想情況下，我想用meanTraps爲每個站點的數據幀。 任何人都可以幫助糾正我的錯誤？

Answer 1

stat<-function(x,i) c(m1=mean(x$TR[i])) 
library(boot) 
Trap=dlply(df,.(df$Sites),function(dat) boot(dat,stat,R=10)) 
ldply(Trap,.fun=function(x) data.frame(mean=mean(x$t,std.err=sd(x$t)))