I have this code that produces a data frame in R (corresponding to repetitions of a random dice rolling experiment until two consecutive 6’s are obtained – source: R: Recursively checking dice combinations):

sides <- 1:6
rolls <- sample(sides, size = 100, replace = TRUE)
counts <- c()
position <- 1

while (position < length(rolls)) {
    # check if the current and next roll are both 6
    if (rolls[position] == 6 & rolls[position + 1] == 6) {
        
        counts <- c(counts, position)
        #move to the next position 
        position <- position + 2
    } else {
       
        position <- position + 1
    }
}

# result data frame
result <- data.frame(current_occurrence = integer(), absolute_position = integer(), positions_away_from_current_occurence = integer(), six_six = character())

# Loop over all occurrences of (6,6)
for (i in 1:(length(counts)-1)) {
    # Get the current occurrence
    current_occurrence <- counts[i]
    
    # Initialize j
    j <- 2
    
    # Loop until we reach the next occurrence of (6,6) or the end of the rolls
    while ((current_occurrence + j <= counts[i+1])) {
        # Check if (6,6) occurred at (current_occurrence + j - 1, current_occurrence + j)
        six_six <- if (rolls[current_occurrence + j - 1] == 6 & rolls[current_occurrence + j] == 6) "yes" else "no"
        
        # add the result to the data frame
        result <- rbind(result, data.frame(current_occurrence = i, absolute_position = current_occurrence, positions_away_from_current_occurence = j, six_six = six_six))
        
    
        j <- j + 1
    }
}

The results look like this:

print(result)

   current_occurrence absolute_position positions_away_from_current_occurence six_six
1                   1                 4                                     2      no
2                   1                 4                                     3      no
3                   1                 4                                     4      no
4                   1                 4                                     5      no
5                   2                 9                                     2     yes
6                   2                 9                                     3      no
7                   2                 9                                     4      no
8                   2                 9                                     5      no
9                   2                 9                                     6      no
10                  2                 9                                     7      no

I am trying to accomplish the following task:

• Isolate all rows where current_occurence=1 . Call the first row as b*
• Check whether six_six=yes at (b*, b* + 1)
• Check whether six_yes = yes at (b*, b* +2)
• Repeat this process until you reach the last row in current_occurence=1
• Repeat this process for all values of current_occurence

In the end I am trying to produce a data frame that looks like this:

#sample result of desired output:

   current_occurence b_star after_b_star six_six
                  1      1            1      no
                  1      1            2      no
                  1      1            3      no
                  1      1            4      no
                  1      1            5     yes
                  2      1            1      no
                  2      1            2      no
                  2      1            3      no
                  2      1            4      no
                  2      1            5      no

My friend and I spent some time yesterday trying to write the code for this without getting confused within the loops and subloops:

desired_result <- data.frame(current_occurrence = integer(), b_star = integer(), after_b_star = integer(), six_six = character())

# loop over all unique occurrences
for (current_occurrence in unique(result$current_occurrence)) {
    # subset for the current occurrence
    subset_result <- result[result$current_occurrence == current_occurrence, ]
    
   
    b_star <- subset_result[1, ]
    
    # all rows after b_star
    for (i in 2:nrow(subset_result)) {
        # check if six_six is "yes" at (b_star, b_star + i)
        six_six <- if (subset_result[i, "six_six"] == "yes") "yes" else "no"
        
        # append
        desired_result <- rbind(desired_result, data.frame(current_occurrence = current_occurrence, b_star = 1, after_b_star = i - 1, six_six = six_six))
    }
}

print(desired_result)

The output is in the correct format, but we are not sure the code is doing what we want it to do:

    current_occurrence b_star after_b_star six_six
                    1      1            1      no
                    1      1            2      no
                    1      1            3      no
                    2      1            1      no
                    2      1            2      no
                    2      1            3      no
                    2      1            4      no

Is there a standard way to do these kinds of recursive tasks? Can someone please help?

Thanks!