How to get the cumulative distribution function with NumPy?
Using a histogram is one solution but it involves binning the data. This is not necessary for plotting a CDF of empirical data. Let F(x)
be the count of how many entries are less than x
then it goes up by one, exactly where we see a measurement. Thus, if we sort our samples then at each point we increment the count by one (or the fraction by 1/N) and plot one against the other we will see the "exact" (i.e. un-binned) empirical CDF.
A following code sample demonstrates the method
import numpy as npimport matplotlib.pyplot as pltN = 100Z = np.random.normal(size = N)# method 1H,X1 = np.histogram( Z, bins = 10, normed = True )dx = X1[1] - X1[0]F1 = np.cumsum(H)*dx#method 2X2 = np.sort(Z)F2 = np.array(range(N))/float(N)plt.plot(X1[1:], F1)plt.plot(X2, F2)plt.show()
It outputs the following
I'm not really sure what your code is doing, but if you have hist
and bin_edges
arrays returned by numpy.histogram
you can use numpy.cumsum
to generate a cumulative sum of the histogram contents.
>>> import numpy as np>>> hist, bin_edges = np.histogram(np.random.randint(0,10,100), normed=True)>>> bin_edgesarray([ 0. , 0.9, 1.8, 2.7, 3.6, 4.5, 5.4, 6.3, 7.2, 8.1, 9. ])>>> histarray([ 0.14444444, 0.11111111, 0.11111111, 0.1 , 0.1 , 0.14444444, 0.14444444, 0.08888889, 0.03333333, 0.13333333])>>> np.cumsum(hist)array([ 0.14444444, 0.25555556, 0.36666667, 0.46666667, 0.56666667, 0.71111111, 0.85555556, 0.94444444, 0.97777778, 1.11111111])
update for numpy version 1.9.0. user545424's answer does not work in 1.9.0. This works:
>>> import numpy as np>>> arr = np.random.randint(0,10,100)>>> hist, bin_edges = np.histogram(arr, density=True)>>> hist = array([ 0.16666667, 0.15555556, 0.15555556, 0.05555556, 0.08888889, 0.08888889, 0.07777778, 0.04444444, 0.18888889, 0.08888889])>>> histarray([ 0.1 , 0.11111111, 0.11111111, 0.08888889, 0.08888889, 0.15555556, 0.11111111, 0.13333333, 0.1 , 0.11111111])>>> bin_edgesarray([ 0. , 0.9, 1.8, 2.7, 3.6, 4.5, 5.4, 6.3, 7.2, 8.1, 9. ])>>> np.diff(bin_edges)array([ 0.9, 0.9, 0.9, 0.9, 0.9, 0.9, 0.9, 0.9, 0.9, 0.9])>>> np.diff(bin_edges)*histarray([ 0.09, 0.1 , 0.1 , 0.08, 0.08, 0.14, 0.1 , 0.12, 0.09, 0.1 ])>>> cdf = np.cumsum(hist*np.diff(bin_edges))>>> cdfarray([ 0.15, 0.29, 0.43, 0.48, 0.56, 0.64, 0.71, 0.75, 0.92, 1. ])>>>