Yes thanks for that link and the software. Will certainly look into porting it into Python.

One final question. From this API - https://docs.scipy.org/doc/scipy/reference/signal.windows.html would a Welch window be a welcome addition to that API ? I mean the Bartlett one is basically the precursor of the Welch window. Seems odd that a Welch window is missing. 

On Tue, May 17, 2022 at 8:35 PM Robert Kern <robert.kern@gmail.com> wrote:
On Mon, May 16, 2022 at 11:04 AM Robert Kern <robert.kern@gmail.com> wrote:
I don't know of any work that apply's something like Welch's method per se to the LS periodogram, but if you squint, you can kind of see Welch's method as an ad hoc instance of a multitaper method, and there is a more rigorous application of multitaper methods to the LS periodogram. Not implemented in Python so far as I know, but there is an R implementation.


Looking through this paper more carefully, I see it cites this paper that implements multitaper spectral estimate on regularly-sampled but gappy time series like yours.

  https://core.ac.uk/download/pdf/237709478.pdf

The corresponding MATLAB code is BSD-licensed:

  https://www.mathworks.com/matlabcentral/fileexchange/71909-mdmwps

--
Robert Kern
_______________________________________________
SciPy-User mailing list -- scipy-user@python.org
To unsubscribe send an email to scipy-user-leave@python.org
https://mail.python.org/mailman3/lists/scipy-user.python.org/
Member address: winash12@gmail.com