从n组数字中生成x个随机数

Question

Assuming I have the following lists of numbers in R, how can I choose 6 random numbers by specifying the how many I need from each group?

r<-c(1,2,3,4,5,6,7,8,9,10)
b<-c(11,12,13,14,15,16,17,18,19,20)
y<-c(21,22,23,24,25,26,27,28,29,30)
g<-c(31,32,33,34,35,36,37,38,39,40)

How can I select, say 2 random numbers from r, 2 random numbers from b, 1 random number from y and 1 random number from g?

I need my codes to be flexible enough to manually input how many numbers I need from each list. I also want the final output to be just the list of 6 random numbers, based on my selection.

Answer 1

Put vectors into one list

foo <- list(
    c(1,2,3,4,5,6,7,8,9,10),
    c(11,12,13,14,15,16,17,18,19,20),
    c(21,22,23,24,25,26,27,28,29,30),
    c(31,32,33,34,35,36,37,38,39,40)
)

---

Sample x times

If you don't mind replacement then generate another vector of draws (specifies from which list entry to sample) and iterate over this vector.

bar <- c(1, 1, 2, 2, 3, 4)
sapply(bar, function(x) sample(foo[[x]], 1))

---

Sample n times

If you don't want replacement then you have to sample only once from each list entry. To do this you have to name your list entries with how many draws you want and iterate over the list (for each entry extract name and sample this number).

names(foo) <- c(2, 2, 1, 1)
unlist(sapply(seq_along(foo), function(x) sample(foo[[x]], names(foo[x]))))

---

Sample 1 time

Another solution is to generate vector of probabilities and pass it as prob argument in sample .

sample(unlist(foo), length(bar),
       prob = rep(table(bar) / length(bar) / lengths(foo), lengths(foo)))

Explanation:

How many samples from each entry we need: table(bar)

1 2 3 4 
2 2 1 1 

Assign each entry a probability: table(bar) / length(bar)

        1         2         3         4 
0.3333333 0.3333333 0.1666667 0.1666667 

Give each number it's probability to be sampled: rep(table(bar) / length(bar) / lengths(foo), lengths(foo))

...

         2          2          2          2          2          2          2          2          2          2 
0.03333333 0.03333333 0.03333333 0.03333333 0.03333333 0.03333333 0.03333333 0.03333333 0.03333333 0.03333333 
         3          3          3          3          3          3          3          3          3          3 
0.01666667 0.01666667 0.01666667 0.01666667 0.01666667 0.01666667 0.01666667 0.01666667 0.01666667 0.01666667 

...