Discover MakerZone

MATLAB and Simulink resources for Arduino, LEGO, and Raspberry Pi

Learn more

Discover what MATLAB® can do for your career.

Opportunities for recent engineering grads.

Apply Today

Thread Subject:
significance testing of power spectra

Subject: significance testing of power spectra

From: Florian Maurer

Date: 13 Mar, 2002 18:48:20

Message: 1 of 2

Some questions about testing the significance of peaks in power spectra,
which should be for interest for many people dealing with it:

1) normal periodogram - conventional FFT
Assume plotting the real positive values of the FFT of a time series
versus the frequencies. This gives a power spectrum, also called
periodogram. How can I test the significance of the various peaks? How
do I have to calculate the upper and lower confidence levels of the
signals, when I want to compare them with the red noise of the data?
Does somebody have an m-file for that?

2) PMTM - multitaper method
I created some power spectra for different time series using the
multitaper method (with m-file PMTM). I know that the significance of
the various signals in the spectra can be tested by plotting the lower
confidence level against the red noise of the data.
Does somebody have a script that can perform the F-test of these spectra
condisering the algorithms given in the pmtm-calculations?

Maybe some of the Mathworks programmers have an answer for the second
question!?

Many regards


--
Posted via Mailgate.ORG Server - http://www.Mailgate.ORG

Subject: significance testing of power spectra

From: Roman Kovalenko

Date: 14 Mar, 2002 10:12:08

Message: 2 of 2

Florian Maurer wrote:
>
>
> Some questions about testing the significance of peaks in
> power spectra,
> which should be for interest for many people dealing with
> it:
>
> 1) normal periodogram - conventional FFT
> Assume plotting the real positive values of the FFT of a
> time series
> versus the frequencies. This gives a power spectrum, also
> called
> periodogram. How can I test the significance of the various
> peaks? How
> do I have to calculate the upper and lower confidence
> levels of the
> signals, when I want to compare them with the red noise of
> the data?
> Does somebody have an m-file for that?
>
> 2) PMTM - multitaper method
> I created some power spectra for different time series
> using the
> multitaper method (with m-file PMTM). I know that the
> significance of
> the various signals in the spectra can be tested by
> plotting the lower
> confidence level against the red noise of the data.
> Does somebody have a script that can perform the F-test of
> these spectra
> condisering the algorithms given in the pmtm-calculations?
>
> Maybe some of the Mathworks programmers have an answer for
> the second
> question!?
>
> Many regards
>
>
> --
> Posted via Mailgate.ORG Server - http://www.Mailgate.ORG <http://www.Mailgate.ORG>
> Are you interested in color noises?
Please tell me by email.
I'm interesting in spectral estimation of signals, corrupted by
noises...
Best regards,
Roman.

Tags for this Thread

No tags are associated with this thread.

What are tags?

A tag is like a keyword or category label associated with each thread. Tags make it easier for you to find threads of interest.

Anyone can tag a thread. Tags are public and visible to everyone.

Contact us