4c0e4c89$1@news.povray.org...
>> OK. I'll go see if I can get that to work...
>
> Right, that seems to work quite well...
>
> Any idea how to get a confidence estimate?

Years ago I wrote some VBA code to calculate regressions with all 
parameters. The code was adapted from pseudocode found in books of 
statistics, so parts of it are pure magic: I don't understand how it works. 
I can send it to you if you want.

G.

Post a reply to this message