Prior work has shown that causal structure can be uniquely identified from observational data when these follow a structural equation model whose error terms have equal variances. We show that this fact is implied by an ordering among (conditional) variances. We demonstrate that ordering estimates of these variances yields a simple yet state-of-the-art method for causal structure learning that is readily extendable to high-dimensional problems.
This repository maintains the code for this project.
Low dimensional methods are in the files EqVarDAG_TD.R and EqVarDAG_BU.R.
High dimensional methods are in the files EqVarDAG_HD_TD.R and EqVarDAG_HD_CLIME.R.
The methods can be called in the format EqVarDAG_TD(X) for the n-by-p data matrix X.
Simulation studies as described in the paper can be re-produced from the R files in the Experiment folder.