[Gnuplot-info] Re: Question about Gnuplot

[K.Moriyama <km...@mb...>]

Dear Riccardo,

You can define the functions so that they have the
same value at x=5.
For example, define 
f1(x)= a*(x-5)**2+b*(x-5)+c ---> f1=c at x=5
f2(x)= d*(x-5)**2+e*(x-5)+c ---> f2=c at x=5, too!
then, fit f1(x) for data set 1 and you will get a,b,c.
Using same c, you can fit f2 for data set 2 and get d and e.
Maybe, you can fit c for data set 2 again and 
do the f1 with the new c again.
Iterate the procedure until (hopefully) you get a converged value for c.
I do not know if c really converges, though ;-|

mori/Japan

"Riccardo S." <ric...@ya...> wrote
on Thu, 24 Feb 2005 19:28:45 +0100 (CET)
Subj.: [Gnuplot-info] Question about Gnuplot
> My name is Riccardo, and I'm a physics student from
> Italy.
> I have a question about a function of Gnuplot.
> I have two curves, each of them is a fit of
> experimental data, in two different ranges (let's
> say[0:5] and [5:10]): how can I make them to pass thru
> a same point in x=5?
> I tried to find the answer in the handbook, but
> didn't.
> Thank you in advance
> Riccardo Senica