I am trying to create a dataframe in R.

I have 4 categories (e.g. Blue, Red, Yellow, Green) and I would like each row to sum to 100%. For each category I want to create incrimental differences of 5% units and produce a dataframe which has all possible combinations of numbers (to nearest 5%) for the 4 categories. I realise I am not explaining this well at all, so I have tried to show what I mean in the following table:

I am struggling to know where to start here...

CodePudding user response：

You could nest three for loops and bind the results together:

target_df <- data.frame()

for (i in seq(95, 0, by = -5)) {
  for (j in seq(100 - i, 0, by = -5)) {
    for(k in seq(100 - i - j, 0, by = -5)) {
      target_df <- rbind(target_df, data.frame(Blue = i, Red = j, Yellow = k, Green = 100 - i - j - k))
    }
  }
}

This returns

    Blue Red Yellow Green
1     95   5      0     0
2     95   0      5     0
3     95   0      0     5
4     90  10      0     0
5     90   5      5     0
6     90   5      0     5
7     90   0     10     0
8     90   0      5     5
9     90   0      0    10
10    85  15      0     0

You might want to remove three rows containing 100 in columns Red, Yellow and Green.