Crude periodicity finding

Discover the periods in evolution of animal populations (../../data/populations.txt)

Load the data

import numpy as np
data = np.loadtxt('../../../../data/populations.txt')
years = data[:, 0]
populations = data[:, 1:]

Plot the data

import matplotlib.pyplot as plt
plt.plot(years, populations * 1e-3)
plt.ylabel('Population number ($\cdot10^3$)')
plt.legend(['hare', 'lynx', 'carrot'], loc=1)

Plot its periods

from scipy import fftpack
ft_populations = fftpack.fft(populations, axis=0)
frequencies = fftpack.fftfreq(populations.shape[0], years[1] - years[0])
periods = 1 / frequencies
plt.plot(periods, abs(ft_populations) * 1e-3, 'o')
plt.xlim(0, 22)
plt.ylabel('Power ($\cdot10^3$)')

There’s probably a period of around 10 years (obvious from the plot), but for this crude a method, there’s not enough data to say much more.

Total running time of the script: ( 0 minutes 0.150 seconds)

Gallery generated by Sphinx-Gallery