R pass value of column as value for selecting column from array

I am trying to pass the value of an existing R attribute (column) as the value needed for identifying the particular column in an array I want to select. I previously wrote an For loop with an IF statement but it is running really slow.

Each person will have a group membership (1, 2, or 3). I have the probability of their group membership in data$prob1, data$prob2, and data$prob3.

I want to pass the value of data$Group as follows:

data$ClstrAffinity = data$Prob[ , data$Group]

but it does not work. Any ideas?

The slow running code is below.

Thank you.

data$ProbOne = data$Prob[ , 1]
data$ProbTwo = data$Prob[ ,2]
data$ProbThree = data$Prob[ ,3]
data$GroupMembershipNumeric = as.numeric(data$Group)

data[data$Group == 1]

for (a in c(1:nrow开发者_Python百科(data))) {
  groupMembership = data$GroupMembershipNumeric[a]
  if (groupMembership == 1) {
    data$ClstrAffinity[a] = data$ProbOne[a]
  }
  if (groupMembership == 2) {
    data$ClstrAffinity[a] = data$ProbTwo[a]
  }
  if (groupMembership == 3) {
    data$ClstrAffinity[a] = data$ProbThree[a]
  }
  print(groupMembership)
  groupMembership = NULL
}

Matrix indexing is what you need. I'll generate some sample data

set.seed(5)
Prob <- matrix(sample(0:10, 15, replace=TRUE)/10, ncol=3)
Group <- sample(1:3,5,replace=TRUE)

Then your desired result is just

ProbFinal <- Prob[cbind(1:5,Group)]

You may use which:

data$ProbFinal = "NA"

data$ProbOne = data$Prob[ ,1]
data$ProbTwo = data$Prob[ ,2]
data$ProbThree = data$Prob[ ,3]

data$ProbFinal[which(data$Group == 1)] = data$ProbOne[which(data$Group  == 1)]
data$ProbFinal[which(data$Group == 2)] = data$ProbTwo[which(data$Group == 2)]
data$ProbFinal[which(data$Group == 3)] = data$ProbThree[which(data$Group == 3)]