On 18 August 2011 03:07, Eric Keller <[email protected]> wrote: > You can get bogged down for years in this subject. When I want to do > something quick and dirty, I use a least squares filter.
Is this the same as the approach that I (re?)invented for deriving velocity from noisy data: 1) Fit a least-squares polynomial to the data raw_data -> a + bx + cx^2 + dx^3... 2) Take the explicit differential of this polynomial gradient = b + 2cx + 3dx^2... 3) Solve for the required X (potentially in the future, with caveats) I have (almost undocumented, and forgotten) code somewhere that does the same thing in up to 7 linear dimensions, if it is any use to anyone. -- atp "Torque wrenches are for the obedience of fools and the guidance of wise men" ------------------------------------------------------------------------------ Get a FREE DOWNLOAD! and learn more about uberSVN rich system, user administration capabilities and model configuration. Take the hassle out of deploying and managing Subversion and the tools developers use with it. http://p.sf.net/sfu/wandisco-d2d-2 _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
