Re: [Gnuplot-info] fitting polynomial function

[Hans-Bernhard B. <HBB...@t-...>]

Becker, Martin wrote:

> But when I try to evaluate the polynomial function (typing in the 
> "final set of parameters" I get from gnuplot manually), 

Why on earth would you do that?  The results have all been stored into 
the respective user-defined variables under the names you gave them.

Try "show var" after you fit and you'll see.  You'll also want to 
re-read the "help fit" pages.

> a               = -20719.6         +/- 4.261e+04    (205.6%)
> b               = 186386           +/- 3.689e+05    (197.9%)
[...]

That's not a good fit by any stretch of the imagination.  All resulting 
parameters are completely dominated by error.

> The problem might be that the coefficients I get from gnuplot's fit
> function are lacking some decimals, but I'm not sure about that.

That's exactly the problem.  It would be utterly pointless to print all 
15 decimals in each of those parameters in a fit result summary. 
Especially in a fit like this, where not even the first digit is 
actually significant!