Question

我想要制作400个样本，其中每个样本包含来自我的数据集的50个值。我想在for-loop中进行，但我不知道该怎么做。以下是我目前的代码

 set.seed(1)
 alldata     = humangenes$GC ### I am storing the value of GC variable 
 truemean    = mean(alldata)
 nsamples    = 400   ###number of samples I want to make
 r = data.frame()

  for (j in 1:nsamples) {
        mysample = sample(alldata, 50, replace = F) # Each gene can  
        only be picked once

  ##I don't know what to do next
   }

建议将不胜感激。

Answer 1

如果你想使用一个循环，你可以获得像这样的矩阵结果

set.seed(1)
alldata <- rnorm(1000, 1,5) # create sample data
nsamples <- 400             #number of samples I want to make
sampSize <- 50              #size of each sample    

r = <- matrix(nrow = nsamples, ncol = sampSize)

for (j in 1:nsamples) {
     r[j,] <- sample(alldata, sampSize, replace = FALSE)
}

然后可以通过执行

将其转换为数据帧

df <- data.frame(r)

正如Zheyuan在评论中指出，循环可以替换为

r <- replicate(nsamples, sample(alldata, sampSize, replace = F))

如何制作50个值的400个样本

1 个答案: