Search All of the Math Forum:
Views expressed in these public forums are not endorsed by
NCTM or The Math Forum.


Luca
Posts:
77
Registered:
6/6/12


Re: Best way to "classify" a vector? ("averaging" subsampling)
Posted:
Mar 4, 2013 4:51 PM


"someone" wrote in message <kh33p0$mhv$1@newscl01ah.mathworks.com>...
> % This sounds like you are trying to compute a "running average" of the vector. > % If so, see the Examples section of > > doc filter
It's not computing a running average. (Well... actually if I computed a running average with a window of 100, in this example, and _then_ pick one point out of 100 the result would be the same, numerically speaking. But the concept would be different.) Actually this was a simplified example for a more complex thing I have to program. What if the values in "t" were not uniformily spaced? Then the running average would not be appliable.
(and BTW... what would be the practical differences between using filter and using conv with an appropriate kernel?)



