Compute Empirical CDF in Matplotlib

empirical_cdf.py

import numpy as np
import seaborn as sns
import matplotlib.pylab as plt
from numpy.random import weibull
from scipy.stats import cumfreq

Nsamples = 5000
Nsamples_new = 500

# generate a weibull distribution
k=5
z = weibull(k,(Nsamples,))

# useful function
def find_nearest_index(vec,val):
    ix = (np.abs(vec-val)).argmin()
    return ix

# nice trick for getting the empirical cdf:
# http://stackoverflow.com/questions/3209362/how-to-plot-empirical-cdf-in-matplotlib-in-python
zsrt = np.sort(z)
cdf = np.arange(len(zsrt))/float(len(zsrt))

normed = np.array([ cdf[find_nearest_index(zsrt,zz)] for zz in z])
print normed