[Scilab-users] polyfit for Scilab

[A Khorshidi]

Hi; 


Dang, Christophe wrote
> In fact, you don't need a specific function as polynomial fitting is just
> a multiple linear regression considering each x^n is a variable[..]

I think it's not really so bad if we have such a function, is it? 


David Chèze wrote
>  you could use the polyfit function that is made available in the stixbox
> module: it is equivalent to matlab's polyfit.
> http://atoms.scilab.org/toolboxes/stixbox/2.2/

And, besides the polyfit() from Stixbox, there is another function with the
same name in Fitters toolbox (I more prefer this one.)  
see: https://atoms.scilab.org/toolboxes/fitters 
Last year when I was writing a tutorial about polynomials handling (in
Persian), I use the following example to show how one can use this function
for curve fitting.  

x=[1 2 3 4 5]; 
y=[5.5 43.1 128 290.7 498.4]; 
exec('polyfit.sci',-1) 
p=polyfit(x,y,3) 
xp=-.5:.1:5.5; 
yp=horner(p,xp); 
plot(x,y,'o',xp,yp), set(gca(),'grid',[6 6]*color('blue'))


So I suggest Scilab Team to add ployfit() as a core Scilab function.

Mersi, 
Mehran
_




--
View this message in context: http://mailinglists.scilab.org/polyfit-for-Scilab-tp4030877p4030898.html
Sent from the Scilab users - Mailing Lists Archives mailing list archive at Nabble.com.