Moritz-Alexander-Kern 060264eb51 add Multitaper PSD test data 2 yıl önce
..
data 060264eb51 add Multitaper PSD test data 2 yıl önce
scripts 060264eb51 add Multitaper PSD test data 2 yıl önce
LICENSE.md 060264eb51 add Multitaper PSD test data 2 yıl önce
README.md 060264eb51 add Multitaper PSD test data 2 yıl önce
authors 060264eb51 add Multitaper PSD test data 2 yıl önce

README.md

Multitaper PSD test data

This folder contains the data used for Multitaper PSD unit tests. Specifically, it contains the MATLAB [Registered Trademark] Multitaper and Nitime Multitaper PSD estimate as well as the ground truth PSD of time series data generated using an autoregressive model (ARM).

Description

scripts/

  • LICENSE (BSD 3-clause)
  • authors -- list of authors and contributors
  • generate_test_data.py -- Python script that generates time series data using ARM, computes the ground truth PSD and estimates its PSD using the Nitime multitaper method [1].
  • matlab_multitaper.m -- MATLAB script that loads the generated time_series.mat and estimates its PSD via pmtm [2] (i.e. Thomson Multitaper method [3]).
  • requirements.txt -- requirements containing the exact package versions used in generate_test_data.py to generate time_series.npy, time_series.mat, psd_groundtruth.npy and psd_nitime.npy in data/ folder. The requirements file was generated using pipreqs-0.4.11 package.

data/

  • LICENSE (CC-BY)
  • time_series.npy -- artificial data generated using ARM (in NumPy format)
  • time_series.mat -- artificial data generated using ARM (in MATLAB format)
  • matlab_multitaper.mat -- data resulting from the matlab_multitaper.m
  • psd_groundtruth.npy -- ground truth PSD of the time series data
  • psd_nitime.npy -- data of multitaper PSD from the Nitime package [1]

Copyright

:copyright: 2021 by the development team.

References and links

  1. http://nipy.org/nitime/examples/multi_taper_spectral_estimation.html
  2. https://de.mathworks.com/help/signal/ref/pmtm.html
  3. Thomson, D. J. (1982) "Spectrum estimation and harmonic analysis." Proceedings of the IEEE, 70, 1055–1096