mca 1.0.4

Multiple correspondence analysis with pandas

mca is a Multiple Correspondence Analysis (MCA) package for python, intended to be used with pandas. MCA is a feature extraction method; essentially PCA for categorical variables. You can use it, for example, to address multicollinearity or the curse of dimensionality with big categorical variables.

Installation

pip install --user mca

Usage

Please refer to the usage notes and this illustrated ipython notebook.

References

• Michael Greenacre, Jörg Blasius (2006). Multiple Correspondence Analysis and Related Methods, CRC Press. ISBN 1584886285.

• François Husson, Multiple Correspondence Analysis Youtube Playlist, Youtube

History

• 1.0 (2014-06-24)

  First release. I’m sure it’s an auspicious date somewhere in the world.

• 1.01 (2015-03-23)

  More documentation, in the form of an ipython notebook. Fixed bug #2 affecting python 2.x

• 1.02 (2017-07-29)

  Fixed division-by-zero bug (issue #14)

• 1.03 (2018-01-10)

  Added sparse matrix support

• 1.04 (2025-05-15)

  Improved SVD efficiency (issue #23)