Uncertainty in Artificial Intelligence
First Name   Last Name   Password   Forgot Password   Log in!
    Proceedings         Authors   Author's Info   Article details         Search    
Testing whether linear equations are causal: A free probability theory approach
Jakob Zscheischler, Dominik Janzing, Kun Zhang
Abstract:
We propose a method that infers whether linear relations between two high-dimensional variables X and Y are due to a causal influence from X to Y or from Y to X. The earlier proposed so-called Trace Method is extended to the regime where the dimension of the observed variables exceeds the sample size. Based on previous work, we postulate conditions that characterize a causal relation between X and Y. Moreover, we describe a statistical test and argue that both causal directions are typically rejected if there is a common cause. A full theoretical analysis is presented for the deterministic case but our approach seems to be valid for the noisy case, too, for which we additionally present an approach based on a sparsity constraint. The discussed method yields promising results for both simulated and real world data.
Keywords:
Pages: 839-846
PS Link:
PDF Link: /papers/11/p839-zscheischler.pdf
BibTex:
@INPROCEEDINGS{Zscheischler11,
AUTHOR = "Jakob Zscheischler and Dominik Janzing and Kun Zhang",
TITLE = "Testing whether linear equations are causal: A free probability theory approach",
BOOKTITLE = "Proceedings of the Twenty-Seventh Conference Annual Conference on Uncertainty in Artificial Intelligence (UAI-11)",
PUBLISHER = "AUAI Press",
ADDRESS = "Corvallis, Oregon",
YEAR = "2011",
PAGES = "839--846"
}


hosted by DSL   •   site info   •   help