Pairwise LvLiNGAM (Pairwise Latent Variable Linear Non-Gaussian Acyclic Model)
On this page you can find matlab code for the pairwise LvLiNGAM algorithm to learn partial information about pairwise total causal effects in a linear non-Gaussian acyclic model. The method is introduced in the paper:
D. Entner, and P. O. Hoyer (2011). Discovering Unconfounded Causal Relationships Using Linear Non-Gaussian Models [pdf]
New Frontiers in Artificial Intelligence, LNAI 6797
(The original publication is available at www.springerlink.com.)
Software
The matlab code of the Pairwise LvLiNGAM algorithm to replicate the results in the above article as well as to apply the method to own data can be downloaded here (and the README file for convenience).
Additionally, the following matlab code packages are needed:
- LiNGAM (Linear Non-Gaussian Acyclic Model) LiNGAM homepage (Direct link to download the codepackage: download)
- lvLiNGAM (latent variable Linear Non-Gaussian Acyclic Model) LiNGAM homepage (Direct link to download the codepackage: download)
- HSIC (Hilbert-Schmidt Independence Criterion; A Kernel Statistical Test of Independence) HSIC homepage (Direct link to download the codepackage: download)
Contact Information
If you have any questions/comments, please contact Doris Entner.
Last update: 29 December 2011