[SciPy-user] Fit a function with errors

Alexander Michael lxander.m@gmail....
Thu Aug 30 10:20:24 CDT 2007


On 8/30/07, Wolfgang Kerzendorf <wkerzendorf@googlemail.com> wrote:
> I am trying to fit a function and get errors for the fit parameters a
> and b:
> a simple example would be: y=a*x+b. I dont want to get only the rms but
> I want to get errors for the parameters a and b. neither linregress nor
> lstsq method do that, I also want to fit x's and y's that have errors
> and then get the appropriate error for a and b.
> To complicate matters;-) :
> I want to fit a function
> : u=a*v+b*w+c*x+d*y +e*z and want to fit the parameters a,b,c,d,e and
> know their errors to values u,v,w,x,y,z in the case of them having
> errors and them having not errors.
> I know this is a lot, but thanks in advance for help.

Have you looked at the scipy cookbooks example OLS?

<http://www.scipy.org/Cookbook/OLS>


More information about the SciPy-user mailing list