How can I generate Power Spectra?

Asked May 20 '16 at 03:50

Active May 20 '16 at 03:50

Viewed 141 times

I have some epidemiological data. The data is inherently periodic, and so I would like to visualize the power spectra so as to gain insight into the frequency of oscillations.

The book I am reading outlines a method to visualize the power spectra:

De-trend the data
Compute the fourier transform of the autocorrelation function of the detrended data.
smooth with suitable spectral window such as Tukey WIndow or Bartlett window.

(Book is Mathematical Epidemiology, page 311-312).

I was wondering if there is a function in python to do this? If not, what functions and libraries can I use?

asked May 20 '16 at 03:50

Demetri Pananos

6,770
9
42
73

1

look into the `scipy` library to find `scipy.fft` – Reblochon Masque May 20 '16 at 03:52
And for the detrending and smoothing with spectral window? – Demetri Pananos May 20 '16 at 03:55
1

numpy has windowing functions to smooth data – Paul H May 20 '16 at 09:48

How can I generate Power Spectra?

0 Answers0