Comparing partitions from split() using a nested for loop containing an if statement-CodePudding

Consider the below MWE that splits a distance matrix and attempts to compare partitions:

set.seed(1234) # set random seed for reproducibility

# generate random normal variates
x <- rnorm(5)
y <- rnorm(5)

df <- data.frame(x, y) # merge vectors into dataframe
d <- dist(x) # generate distance matrix

splt <- split(d, 1:5) # split data with 5 values in each partition

# compare partitions

for (i in 1:length(splt)) {
  for (j in 1:length(splt)) {
     if(splt[[i]] != splt[[j]]) {
      a <- length(which(splt[[i]] >= min(splt[[j]]))) / length(splt[[i]])
      b <- length(which(splt[[j]] <= max(splt[[i]]))) / length(splt[[j]])
    }
  }
}
# Error in if (splt[[i]] != splt[[j]]) { : the condition has length > 1

The above for loop should compare all unique partitions (i.e., (1, 2), (1, 3), ... ,(4, 5)). However, the condition is greater than 1.

The result for comparing partition 1 (split[[1]]) and partition 2 (split[[2]]) for instance should be a = b = 1.

a <- length(which(splt[[1]] >= min(splt[[2]]))) / length(splt[[1]])
b <- length(which(splt[[2]] <= max(splt[[1]]))) / length(splt[[2]])

I know the solution is to instead use ifelse() but there is no else within the nested loop.

Any ideas on how to proceed?

CodePudding user response：

Is your problem the error message? That is, why R does not like your comparison splt[[i]] == splt[[j]]? The reason is that we get a vector of comparisons:

> splt[[1]] != splt[[2]]
[1] TRUE TRUE

If I understand you correctly, splt[[i]] is equal to splt[[j]] if all entries are equal and different otherwise. If so, change the comparison to be !(all(splt[[i]] == splt[[j]])). The total loop looks like this:

for (i in 1:length(splt)) {
    for (j in 1:length(splt)) {
        if (!(all(splt[[i]] == splt[[j]]))) {
            a <- length(which(splt[[i]] >= min(splt[[j]]))) / length(splt[[i]])
            b <- length(which(splt[[j]] <= max(splt[[i]]))) / length(splt[[j]])
        }
    }
}