Talk:Periodogram

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Raw periodogram limitations[edit]

"The raw periodogram is not a good spectral estimate because of spectral bias and the fact that the variance at a given frequency does not decrease as the number of samples used in the computation increases."

  1. spectral bias is unresolved. Is that the same as Spectral leakage?
  2. Why does the variance at a given frequency not decrease as the # of samples increases? —Preceding unsigned comment added by Dfrankow (talkcontribs) 22:33, 18 March 2009 (UTC)[reply]

Needs research[edit]

It would be nice to track down the original source instead of referencing a reference. I unfortunately don't have time to make for this, since I am less interested in mathematics than applications. The Welch method also needs some research. -rs2 17:25, 24 Mar 2004 (UTC)

difference between periodogram and spectrum?[edit]

It would be nice to see an explanation of why the term "periodogram" is used to refer to a technique that, in practice, estimates power density with respect to frequency instead of with respect to period.. 146.6.201.180 (talk) 19:17, 20 January 2010 (UTC)[reply]

It's simply because around the turn of the century, the word for a periodic function was "periodic", not "frequent".....so people searched for "periodicities", not "frequencies". And they still do. Andersen has, indeed, plumped for renaming the periodogram the "spectrogram" but this has not caught on.98.109.227.161 (talk) 20:07, 8 November 2012 (UTC)[reply]

Suggesting that we do not merge[edit]

I think this page could be expanded rather than merged as there is I think enough to produce something on this. (Msrasnw (talk) 10:12, 29 July 2014 (UTC))[reply]

Distinguishing between periodogram analysis and spectral analysis[edit]

I would like to add something like this to the article:

"Spectral analysis is a modernized version of periodogram analysis modified to take account of the stochastic nature of the entire time series, not just the noise component. If it is assumed that economic time series are fully stochastic, it follows that the older periodogram technique is inappropriate and that considerable difficulties in the interpretation of the periodograms of economic series may be encountered" - Francis X. Diebold, Lutz Kilian and Marc Nerlove "time series analysis" From The New Palgrave Dictionary of Economics, Second Edition, 2008 Edited by Steven N. Durlauf and Lawrence E. Blume (2008). The new Palgrave dictionary of economics. Palgrave Macmillan.

But am not sure how it would best be included. (Msrasnw (talk) 10:32, 29 July 2014 (UTC))[reply]

Assessment comment[edit]

The comment(s) below were originally left at Talk:Periodogram/Comments, and are posted here for posterity. Following several discussions in past years, these subpages are now deprecated. The comments may be irrelevant or outdated; if so, please feel free to remove this section.

This definition of the periodogram completely ignores the fact that is is commonly used to describe the process of calculating the power in signal at various frequencies by simply applying the fourier transform. It also bears mentioning that this technique introduces a lot of bias and does not resolve signals which have very narrow peaks.

Last edited at 20:35, 4 February 2007 (UTC). Substituted at 02:39, 30 April 2016 (UTC)