The Savitzky-Golay Filter is a polynomial smoothing filter. This version implements 3rd degree polynomials using coefficients from Savitzky and Golay's table, specifically the coefficients for a 5-, 7-, 9-, 15- and 25-point window moving averages. The filters are offset to the left by the number of coefficients (n-1)/2 so it smooths on top of the actual curve. You can turn off some of the smoothing curves, as it can get cluttered displaying all at once. Any feedback is very welcome.
Note di rilascio
⋅
Savitzky-Golay digital filter, useful for smoothing data. This script implements the 5-, 7-, 9-, 15-, and 25-point window 3rd degree polynomial fit coefficients. The plots are offset by (n-1)/2 where n is the window size, so it fits the original data. Any feedback is welcome.
@RicardoSantos, I've found it very useful to apply it more than once to the same input, although I do not fully understand the consequences of doing so.
One of the nicest uses was to further smooth a simple moving average (which is not far from the savitzky-golay filter): when you take, for example, a derivative of the smoothed SMA, you get much nicer results.