D2C: Predicting Causal Direction from Dependency Features
The relationship between statistical dependency and causality lies at the heart of all statistical approaches to causal inference. The D2C package implements a supervised machine learning approach to infer the existence of a directed causal link between two variables in multivariate settings with n>2 variables. The approach relies on the asymmetry of some conditional (in)dependence relations between the members of the Markov blankets of two variables causally connected. The D2C algorithm predicts the existence of a direct causal link between two variables in a multivariate setting by (i) creating a set of of features of the relationship based on asymmetric descriptors of the multivariate dependency and (ii) using a classifier to learn a mapping between the features and the presence of a causal link
- Gianluca Bontempi, Catharina Olsen, Maxime Flauder
- Date of publication
- 2015-01-21 00:23:55
- Catharina Olsen <email@example.com>
- Alarm dataset
- Balanced Error Rate
- compute N samples according to the network distribution
- An S4 class to store the RF model trained on the basis of the...
- An S4 class to store DAG.network
- Dataset of the Alarm benchmark
- compute descriptor
- stored D2C object
- creation of a D2C.descriptor
- creation of a D2C object which preprocesses the list of DAGs...
- creation of a DAG.network
- creation of a "simulatedDAG" containing a list of DAGs and...
- mIMR (minimum Interaction max Relevance) filter
- predict if there is a connection between node i and node j
- An S4 class to store a list of DAGs and associated...
- Adjacency matrix of the Alarm dataset
- update of a "D2C" with a list of DAGs and associated...
- update of a "simulatedDAG" with a list of DAGs and associated...
Files in this package