Weighted Scatterplot Formula

[rylock]

Hello!

Does anybody know a way to weight the line of best fit towards points on a scatterplot that are 'more important' than others? So, if I'm looking for the 'y = mx+b' formula at the end of the day, I'd like it to, for example, still take 50 out of the 100 points on the scatterplot into account -- but skew the line/formula more heavily towards the other 50 points that I designated with a heavier weight.

Is that possible?

Thanks!

[shg]

I can show you how to do it if you know how to use Solver.

[rylock]

Thanks again, shg -- and I can learn Solver if that's the only way to do it.

[shg]

There's almost surely a way to do it analytically, but I don't know it. So go read about Solver, and then post back.

[rylock]

Cool, thanks again -- I'll check it out further. If, in the mean time, anybody knows how to do it analytically, that'd be great!