[R] Survey of "moving window" statistical functions - still looking f or fast mad function

[Prof Brian Ripley]

On Fri, 8 Oct 2004, Tuszynski, Jaroslaw W. wrote:

[...]

> 	Finally a question: I still need to get moving windows mad function
> faster my "runmad" function is not that much faster than apply/embed combo,
> and that I used before, and this is where my code spends most of its time. I
> need something like "runmed" but for a mad function. Any suggestions?

Write your own C-level implementation, as runmed and most of the other
fast functions you cite are.

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595