Summing Vector elements until a Positive element is encountered - R

Question

I'm new to R and I'm looking through a book called "Discovering Statistics using R". Although the book implies you don't need any statistical background, some of the content isn't covered/explained...

I'm trying to sum the elements of a vector starting from position 1 until a positive element is present.

I found this question which is very similar to what I'm trying to achieve. However when I implement it, it doesn't always seem to work (and it sometimes appears to include the first positive element)...

My program is:

  vecA <- runif(10, -10, 10);
  sumA <-sum(vecA [1:min(which(vecA < 0))]);

Is there a more robust way to calculate this without using loops that works every time and doesn't add the positive element? I'm not at the looping stage of my books yet.

I also found this site which asks a similar question but their answer errors:

  sum(vecA [seq_len(which.max(vecA > 0)]);

What you're doing is perhaps the simplest way. The answers in the comments are good, but over complicated, making it difficult to understand a simple idea. — Drew75, Jan 31 '14 at 18:44
What is your desired output if the first element is positive (there are no elements to sum)? — Blue Magister, Jan 31 '14 at 18:47
@BlueMagister I assume it should return 0? That seems reasonable anyway. — Reanimation, Jan 31 '14 at 18:55

score 3 · Answer 1 · answered Jan 31 '14 at 18:51

3

You can use the following code:

sum(vecA * !cumsum(vecA > 0))

This also works if the first element is positive or all elements are negative.

answered Jan 31 '14 at 18:51

Sven Hohenstein

80,497
17
145
168

Ah. Thanks great. I've not seen the `cumsum()` function before. Thanks for posting. – Reanimation Jan 31 '14 at 18:55
This is a great answer, upboats for your use of ! my friend – Steve Reno Jan 31 '14 at 19:12

score 1 · Accepted Answer · answered Jan 31 '14 at 18:48

1

You want to use > not < to sum all elements until the first positive one is reached.

You're currently summing from 1 until the first negative value is reached (including the first negative value).

sum(vecA[1:min(which(vecA>0))-1])

the which() function will return all of the positions of the positive elements, then taking the sum from 1 to the position of the first positive - 1 will guarantee you are summing all of the negative elements

answered Jan 31 '14 at 18:48

Steve Reno

1,304
3
14
21

Oh! silly mistake. It makes sense now. Thanks for taking the time post. Very good answer. :) – Reanimation Jan 31 '14 at 18:54
Would this also work if all elements are negative like the answer below? – Reanimation Jan 31 '14 at 19:00
@Reanimation No, it does not work if all elements are negative. – Sven Hohenstein Jan 31 '14 at 19:01
Ah that's ok. I was just checking I wasn't doing something stupid in R when I was trying it out. Thanks. – Reanimation Jan 31 '14 at 19:02

score 1 · Answer 3 · answered Jan 31 '14 at 21:48

match function is usually the fastest to find the first occurrence of some element in a vector, so another version of this could look like follows:

first.positive <- match(TRUE, vecA > 0)
sumA <- sum( vecA[ 1 : first.positive ] ) - vecA[first.positive]

This will give you zero if positive element is the first.

Summing Vector elements until a Positive element is encountered - R

3 Answers3