Fit a non parametric copula¶

In this example we are going to estimate a normal copula from a sample using non parametric representations.

[1]:

from __future__ import print_function
import openturns as ot
import matplotlib.pyplot as plt

[2]:

# Create data
R = ot.CorrelationMatrix(2)
R[1, 0] = 0.4
copula = ot.NormalCopula(R)
sample = copula.getSample(30)

[3]:

# Estimate a normal copula using BernsteinCopulaFactory
distribution = ot.BernsteinCopulaFactory().build(sample)

[4]:

# Draw fitted distribution
distribution.drawPDF()

[4]:

[5]:

# Estimate a normal copula using KernelSmoothing
distribution = ot.KernelSmoothing().build(sample).getCopula()
distribution.drawPDF()

[5]:

OpenTURNS