-1
$\begingroup$

The function fft allows one to zero-pad, and mscohere, after all works with auto and cross spectra, which, presumably, have all been calculated with fft. I haven't found any mention on zero-padding in the documentation for mscohere.

$\endgroup$
1
$\begingroup$

have you tried >> type mscohere.m ? It looks like you need to do is modify the call to welch.

The routine doesn't calculate a confidence region so, yes you could formulate a call to a coherence calculation with zero padding but the sample probability distributions published in any of the papers by Cliff Carter, Al Nuttal, and Charlie Knapp would not apply. The biggest error with using mscohere is not averaging enough. Each bin needs around 500 independent samples to have roughly a $\pm$10% accuracy of the full value range of $[0,1]$. Given the independent averaging issue, and that zero padding does nothing for that, it's hard to see a justification.

SNR effects the value of the coherence, not the accuracy of the estimate. It is not a technique suitable to small sample sets.

$\endgroup$
  • $\begingroup$ Thanks for the references. I can see why it's not suitable to small samples. Well, small samples is exactly what I've got (time series spanning 230 to 270 data points) , and indeed I seem to be getting jibberish by way of results. Yet I've essentially been told by a professor to look for the signal-processing analogy of cross correlations, and, as far as I know, this is what I've got to do. Are you aware of any rigourous way of deciding about window length and number of segments relative to sample size? $\endgroup$ – Fede C Oct 8 '18 at 17:48
  • $\begingroup$ can’t see because you haven’t looked. $\endgroup$ – Stanley Pawlukiewicz Oct 8 '18 at 19:07
  • $\begingroup$ Well, I suppose I can, actually. Because the results are clearly rubbish and that tallies up with you saying coherence isn't suitable to small samples, and it's pretty obvious that this is your field. But thanks. $\endgroup$ – Fede C Oct 8 '18 at 22:05
  • $\begingroup$ boot strap is a way to approach small sample sizes. Its Voodoo but no. chickens must die $\endgroup$ – Stanley Pawlukiewicz Oct 8 '18 at 22:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.