Question

I would like to convert this piece of logic with purely matrix operations instead of for loops. The logic is that in my binary value vector, I want to note every transition point (i.e. where 0 turns to 1 and where 1 turns to 0). Otherwise I want to retain the original values. While a simple loop is fast enough for small vector, I’ll need to perform this operation several times over large datasets hence the need for efficiency via matrics.

x <- c(1,0,0,0,0,1,1,0,0,1,0,1,0,1,0,0,0,1);
y <- rep(-2, length(x));
y[1] <- x[1];
for(i in 2:length(x)){
     if((x[i]==1 && x[i-1]==0) || (x[i]==0 && x[i-1]==1)){
       y[i] = -1;
    }
    else{
       y[i] = x[i];
     }
}

final value of y is
1 -1 0 0 0 -1 1 -1 0 -1 -1 -1 -1 -1 -1 0 0 -1

I’m a new convert to R, many thanks in advance

Answer 1

You can use rle for this:

x.rle <- rle(x)$lengths

x[cumsum(x.rle[1:(length(x.rle) - 1)]) + 1] <- -1


> x
 [1]  1 -1  0  0  0 -1  1 -1  0 -1 -1 -1 -1 -1 -1  0  0 -1