CODEWITHSUNDEEP
pythonscipystatisticsdistributionkolmogorov-smirnov
kowser66
kowser66

Reputation: 175

How to plot figure for KS test

enter image description here

I have two arrays with different sizes I can easily find out the KS statistics using scipy.stats.ks_2samp. But how can I draw CDF like the picture?

Upvotes: 2

Views: 1627

Answers (1)

Physics Enthusiast
Physics Enthusiast

Reputation: 211

One big part of this is sns.ecdfplot. In particular see this SO answer:

Code to produce the cumulative distribution plots:

import seaborn as sns
import pandas as pd
sns.ecdfplot(pd.DataFrame({'x': x, 'z': z}))

Edit: It is easier to use np.cumsum like in this SO answer:

As mentioned, cumsum from numpy works well. Make sure that your data is a proper PDF (ie. sums to one), otherwise the CDF won't end at unity as it should. Here is a minimal working example:

import numpy as np from pylab import *

# Create some test data 
dx = 0.01 
X  = np.arange(-2, 2, dx) 
Y  = np.exp(-X ** 2)

# Normalize the data to a proper PDF 
Y /= (dx * Y).sum()

# Compute the CDF 
CY = np.cumsum(Y * dx)

# Plot both 
plot(X, Y) 
plot(X, CY, 'r--')

show()

Upvotes: 0

Related Questions