[Gnuplot-info] Fitting to weighed data points

[Sanaloht <san...@gm...>]

Hi there!

I got a problem and it is not solvable by the fit weight as explained in the
help.
Second, I'm german, that might lead to funny use of expressions. Be creative
in interpreting or ask me. ;)

I got a bunch of data points, which each represents a series of tries:

50 tries with 30 fails is represented by 0.6 (60%) the other parameter of
the point is a sort of percentile but doesn't really matter.

My problem is that not every data point represents the same amount of tries.
Maybe 50 tries lead to one data point and the other data point represents
100 tries.

What I'd like to have is, that, while the least square process, the residue
of every data point is weighed by the amount of tries it represents.

Don't know if GnuPlot can do something like that, but I'm thankful for every
help!

(The weigh explained in the gnuplot help for "fit" just alters the y axis
value, I don't think that solves my problem.)

Another way to solve my problem could be that I make one data point for
every try:
I got 50 tries which lead to one data pint with x axis value of 0.6 and y of
0.4, so I make 50 data points with x/y pairs of 0.6/0.4 
I did that for every data point but that led to very tiny error values. That
is not what I expected.
The multiplying of the same data point doesn't lead to more equations to
solve the equation. So why is that fit more accurate?

So long,

Florida Tom

-- 
View this message in context: http://www.nabble.com/Fitting-to-weighed-data-points-tp19654203p19654203.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.