Copyright (C) 
2010-2011 Jakob Zscheischler
2010-2011 Kun Zhang

This package contains code to the paper
Jakob Zscheischler, Dominik Janzing, Kun Zhang (2011): "Testing whether linear equations are causal: A free probability theory approach".

It is written in Matlab and should work on any machine. 


%%%%%%%%%%%%
LICENSE
%%%%%%%%%%%%

This project is licensed under the (Free)BSD License.


%%%%%%%%%%%%
CODE
%%%%%%%%%%%%

ALWAYS: reows are dimensions, columns are samples

template.m
gives some examples how to apply the code to given data matrices X and Y

highdimmodel.m 		
generates high-dimensional variables [X Y] with the structure X-->Y

conf_model.m 
generates high-dimensional variables [X Y] with the structure X<--Z-->Y

highdimmodel_sparse 
generates high-dimensional variables [X Y] with the structure X-->Y from a sparse transfer matrix A

significance.m
applies the significance test described in the paper

significance_sparse.m
applies the significance test for the dimension reduction approach

twocause.m
calculates the deltas defined in the paper (equation(6))

unimat.m 
calculates uniformly distributed orthogonal matrices

run.m produces Figure 1 of the paper

./sparse
contains code for the dimension reduction approach (preliminary)


%%%%%%%%%%%%%
PROBLEMS
%%%%%%%%%%%%%
If you have problems or questions (or find some bugs!), please do not hesitate to send an email:
jakob.zscheischler ---at--- tuebingen.mpg.de

