
Leveraging directed causal discovery to detect latent common causes
The discovery of causal relationships is a fundamental problem in scienc...
read it

Causal Inference via Kernel Deviance Measures
Discovering the causal structure among a set of variables is a fundament...
read it

Datadriven causal path discovery without prior knowledge  a benchmark study
Causal discovery broadens the inference possibilities, as correlation do...
read it

Causal Mosaic: CauseEffect Inference via Nonlinear ICA and Ensemble Method
We address the problem of distinguishing cause from effect in bivariate ...
read it

A Review on Algorithms for Constraintbased Causal Discovery
Causal discovery studies the problem of mining causal relationships betw...
read it

Causal bootstrapping
To draw scientifically meaningful conclusions and build reliable models ...
read it

Scorebased Causal Learning in Additive Noise Models
Given data sampled from a number of variables, one is often interested i...
read it
Distinguishing cause from effect using observational data: methods and benchmarks
The discovery of causal relationships from purely observational data is a fundamental problem in science. The most elementary form of such a causal discovery problem is to decide whether X causes Y or, alternatively, Y causes X, given joint observations of two variables X, Y. An example is to decide whether altitude causes temperature, or vice versa, given only joint measurements of both variables. Even under the simplifying assumptions of no confounding, no feedback loops, and no selection bias, such bivariate causal discovery problems are challenging. Nevertheless, several approaches for addressing those problems have been proposed in recent years. We review two families of such methods: Additive Noise Methods (ANM) and Information Geometric Causal Inference (IGCI). We present the benchmark CauseEffectPairs that consists of data for 100 different causeeffect pairs selected from 37 datasets from various domains (e.g., meteorology, biology, medicine, engineering, economy, etc.) and motivate our decisions regarding the "ground truth" causal directions of all pairs. We evaluate the performance of several bivariate causal discovery methods on these realworld benchmark data and in addition on artificially simulated data. Our empirical results on realworld data indicate that certain methods are indeed able to distinguish cause from effect using only purely observational data, although more benchmark data would be needed to obtain statistically significant conclusions. One of the best performing methods overall is the additivenoise method originally proposed by Hoyer et al. (2009), which obtains an accuracy of 63+10 0.74+0.05 on the realworld benchmark. As the main theoretical contribution of this work we prove the consistency of that method.
READ FULL TEXT
Comments
There are no comments yet.