copulalib 1.1.0

A library for the copula

==================================================
This library is now part of the ambhas
(http://pypi.python.org/pypi/ambhas/) library.
Updates will only be made in the ambhas library.
==================================================


============
Copulalib
============

This is a package/library in python to model the copulas.
This is being developed under the project `AMBHAS <http://www.ambhas.com/>`_.

This contains module for the following copula
------------------------------------------------
* Frank
* Clayton
* Gumbel


Installing copulalib
======================

Installing copulalib can be done by downloading source file (copulalib--<version>.tar.gz),
and after unpacking issuing the command::

python setup.py install

This requires the usual Distutils options available.

Or, download the copulalib--<version>.tar.gz file and issue the command::

pip /path/to/copulalib--<version>.tar.gz

Or, directly using the pip::

pip install copulalib


Usage
=========
Import required modules::

import numpy as np
import matplotlib.pyplot as plt
from copulalib.copulalib import Copula

Generate random (normal distributed) numbers::

x = np.random.normal(size=100)
y = 2.5*x+ np.random.normal(size=100)

Make the instance of Copula class with x, y and clayton family::

foo = Copula(x, y, family='clayton')

Print the Kendall's rank correlation::

print foo.tau

Print spearmen's correlation::

print foo.sr

Print pearson's correlation::

print foo.pr

Print the parameter (theta) of copula::

print foo.theta

Generate the 1000 samples (U,V) of copula::

X1, Y1 = foo.generate_xy(1000)

For more details see the test.py file inside module.

Changelog
=========

1.1.0 (June 2011)
------------------

- Changed from function orieted to object oriented, documentation improved.

1.0.0 (May 2011)
------------------

- Initial release


Any questions/comments
=========================
If you have any comment/suggestion/question,
please feel free to write me at [e-mail]

So far, the documents are not well explained,
I will be updating them soon.