Identifying events in time series

Hello,

I have a logical vector (made out of 1's and 0's) that corresponds to a time series. For example,

1 1 1 0 1 1 1 0 0 1 1 1 0 0 0 0 0 0

This is just a part of the vector; it has more than 300 000 lines. In this vector I need to identify sequences of 1's and 0's in which there are at least 90% 1's. The sequences should be separated by at least 3 0's.

I found a partial solution of this problem here (<http://stackoverflow.com/questions/3274043/finding-islands-of-zeros-in-a-sequence)>. However, this method does not account for the gaps of single or double 0's.

Any ideas?

Thanks,

DjR

[djr]

I forgot to mention that the minimum length of the sequence should be 48. So, I need sequences with length >= 48 that have at least 90% of 1's.

[djr]

It's OK... I solved it.