Matthew Donadio
The Stat library could really live anywhere. It only contains functions for mean, median, variance, and std. dev.
I think we should have a library of statistics routines somewhere in the "official" hierarchy. I don't think it belongs under DSP, perhaps not even under Numeric? I have a bunch of functions (which I hesitate to call a library) that duplicates SAS's 'uniVar' functionality (what you mention above, as well as skew, kurtosis, and sums of squares) and which also does quantiles and some pretty printing. Ranging from simple to trivial, but well working IMHO, so I'd be happy to submit it if anybody wants it. (I've also got some more stuff for dealing with distributions (e.g. generating randoms from a specific distribution, calculating p-values, and so on), ANOVA and tests, but I haven't fleshed that part out enough for it to be very useful outside my own programs.) -kzm -- If I haven't seen further, it is by standing in the footprints of giants