[SciPy-user] curve_fit step-size and optimal parameters

David Cournapeau cournape@gmail....
Wed Jun 10 11:35:49 CDT 2009

On Thu, Jun 11, 2009 at 1:12 AM, <josef.pktd@gmail.com> wrote:
> On Wed, Jun 10, 2009 at 3:58 AM, Sebastian
> Walter<sebastian.walter@gmail.com> wrote:
>> If you try to fit the frequency with the least-squares distance  the
>> problem is not only nonlinearity
>> but rather the fact that the objective functions has many local minimizers.
>> At least that's what I have observed in a toy example once.
>> Has anyone experience what to do in that case? (Maybe use L1 norm instead?)
> I would look at estimation in frequency domain, which I know next to
> nothing about.

Yes, that's how I would do as well. Use a periodogram with a simple
peak picking algorithm to get the frequency estimate, and then use the
parametrisation suggested by Robert to get A and B (as a linear

If the signal has constant frequency/phase/amplitude with uncorrelated
noise, this should work very well, even in relatively low SNR



More information about the SciPy-user mailing list