# Calculate percentage change within sliding window

i have aggregated minute OHLC data and looking to find large percentage changes of y% within a sliding time window of t for the close value - for example a change of 4% within 15 minutes would be the region i am looking to pick up on.

One idea I'm thinking is taking my minute blocks and building up a larger time slice of 15 minutes from these 1 minute blocks, then that would give me the percentage change from open to close over that time period.

I have found online terms called momentum but that doesn't seem to do what i want - would greatly appreciate pointers and what the above are algorithms called, and any pitfalls in using them that one should be aware (with noisey changes etc)?