Randomly associate elements of two vectors given conditions

Question

I have a data.table of capitals

capitals<-data.table(capital=c(100,50,25,5))
capitals
   capital
1:     100
2:      50
3:      25
4:       5
         


        

Answer 1

Try this for small vectors:

capital=c(100,50,25,5)
loss=c(45,10,5,1)

posC<- order(capital)
posC

lossN <- NULL

for(i in posC){
  temp <- sample(which(loss<=capital[i]),1)
  lossN <- c(lossN, loss[temp])
  loss <-loss[-temp]
}


data.table(capital=capital,loss=lossN[posC])

EDIT

This one is for large vectors:

set.seed(100)
loss=sort(sample(1:5000,100000,replace = T))
capitals = sort(sample(1:100000,100000,replace=T))    

capU <- unique(capitals)
length(capU)

splitLoss <- split(loss,findInterval(loss,sort(c(0,capU))))
head(splitLoss)
splitCap <- split(capitals,findInterval(capitals,sort(c(0,capU))))
head(splitCap)

lossN <- NULL
temp <- NULL

for(i in 1:length(splitLoss)){  
  temp <- c(temp,splitLoss[[i]])  
  for(j in 1:length(splitCap[[i]])){
    id <- sample(1:length(temp),1)
    lossN <- c(lossN, temp[id])
    temp <-temp[-id]      
  }
}

lossN <- c(lossN,ifelse(length(temp)==1,temp,sample(temp)))
data.table(capital=capitals,loss=lossN)

This takes about 7 sec on my machine. The only assumption here is that capitals is sorted and increasing. If needed you can use the order function to make this work for unordered values of capitals in two more lines.

Hope this helps!!