[SciPy-user] Error estimates with leastsq?
Wed Jun 3 21:43:52 CDT 2009
On Wed, Jun 3, 2009 at 10:12 PM, Joseph Smidt <firstname.lastname@example.org> wrote:
> I am trying to best fit data with theory using leastsq. It works,
> in that the best fit curve fits the data fairly well. I was wondering
> how I could find the error bars on the parameters.
> Is this what cov_x is for that leastsq returns? (See
> What is meant by "estimate of the jacobian around the solution"? Is
> this related to the error bars? It says "see curve_fit", but I
> couldn't find that page.
> For example, for output I get the best fit parameters are: [
> 10.8138327 , 25.18203823] with cov_x = [[ 773.42733539,
> [-1791.83769517, 5203.77670479]]
> Is this saying the best fit for parameter 1 is 10.81 +/- sqrt(773)
> and for parameter 2 = 25.18 +/- sqrt(5203)? Thanks.
No, cov_x is not the correct covariance matrix for the parameter
estimates. It needs to be multiplied by the SSE.
curvefit source is here:
>>> from scipy import optimize
You can also look for the discussion 4 months ago (Feb 12 scipy-user)
when curve_fit got introduced and we were checking the correct
More information about the SciPy-user