Check for consecutive values in R

Question

I have data which looks something like the following:

I would like to write a conditional statement which returns TRUE if 3 or more of the 'flag' values consecutively are TRUE, but I'm not sure how to implement this.

I've put in an example data set below:

example <- data.frame(
  value = c(20, 19.5, 14, 14, 25, 19, 23, 25, 30, 33, 38),
  flag = c(F, F,  F, F, T, F, F, T, T, T, T)
)

Answer 1

Two ways, with slightly different results depending on which you need.

1. rle, as already mentioned:

   r <- rle(example$flag)
   r
   # Run Length Encoding
   #   lengths: int [1:4] 4 1 2 4
   #   values : logi [1:4] FALSE FALSE FALSE TRUE
   r$lengths >= 3 & r$values
   # [1] FALSE FALSE FALSE  TRUE
   

   This tells us that the fourth sequence is both TRUE and 3 or longer. We can re-assign to values for those that are true but not long enough with

   r$values <- r$lengths > 3 & r$values
   

   and then invert it:

   example$flag3 <- inverse.rle(r)
   example
   #    value  flag flag3
   # 1   20.0 FALSE FALSE
   # 2   19.5 FALSE FALSE
   # 3   14.0 FALSE FALSE
   # 4   14.0 FALSE FALSE
   # 5   25.0  TRUE FALSE
   # 6   19.0 FALSE FALSE
   # 7   23.0 FALSE FALSE
   # 8   25.0  TRUE  TRUE
   # 9   30.0  TRUE  TRUE
   # 10  33.0  TRUE  TRUE
   # 11  38.0  TRUE  TRUE
   

2. Window operations:

   example$flag3b <- zoo::rollapplyr(example$flag, 3, function(z) length(z) == 3 & all(z), partial = TRUE)
   example
   #    value  flag flag3 flag3b
   # 1   20.0 FALSE FALSE  FALSE
   # 2   19.5 FALSE FALSE  FALSE
   # 3   14.0 FALSE FALSE  FALSE
   # 4   14.0 FALSE FALSE  FALSE
   # 5   25.0  TRUE FALSE  FALSE
   # 6   19.0 FALSE FALSE  FALSE
   # 7   23.0 FALSE FALSE  FALSE
   # 8   25.0  TRUE  TRUE  FALSE
   # 9   30.0  TRUE  TRUE  FALSE
   # 10  33.0  TRUE  TRUE   TRUE
   # 11  38.0  TRUE  TRUE   TRUE