I am dealing with an existing algorithm where data is processed as follows: We have a large sample of data taken every minute, this is grouped into 'buckets' of data for every hour. (Each bucket contains a SUM(), MAX(), MIN() and the number of data points taken in the hour.)
Now, ideally there should have been 60 samples in each bucket, but the algorithm takes the 60th minute data and adds it to current bucket AND the next bucket.
I wanted to know if this is a standard processing technique? Note: The data is prepped for display in a graph later (so is it a smoothing technique?) Note 2: I have very little knowledge of Statistics.