> In this particular application, the shapes are the areas between the > lines created when graphing several time series on a chart (e.g., the > price of several stocks, graphed over time)
It sounds to me like what you need is actually some kind of line of best fit rather than a centre. The usual idea is to find a line (if you do want a point, then this works for a point too) such that the squared distance of all the points in your region from the line (or point) is minimised. You can find this by using a correlation coefficient (a quick Google turned up this page which sums up the process pretty well: http://helios.bto.ed.ac.uk/bto/statistics/tress11.html). For data in more than two dimensions you'll need to use a more complicated process called principal components analysis. Best Danny _______________________________________________ [email protected] To change your subscription options or search the archive: http://chattyfig.figleaf.com/mailman/listinfo/flashcoders Brought to you by Fig Leaf Software Premier Authorized Adobe Consulting and Training http://www.figleaf.com http://training.figleaf.com
