A Scalable Conditional Independence Test for Nonlinear, Non-Gaussian Data

@article{Ramsey2014ASC,
  title={A Scalable Conditional Independence Test for Nonlinear, Non-Gaussian Data},
  author={Joseph Ramsey},
  journal={CoRR},
  year={2014},
  volume={abs/1401.5031}
}
Many relations of scientific interest are nonlinear, and even in linear systems distributions are often non-­‐Gaussian, for example in fMRI BOLD data. A class of search procedures for causal relations in high dimensional data relies on sample derived conditional independence decisions. The most common applications rely on Gaussian tests that can be systematically erroneous in nonlinear non-­‐Gaussian cases. Recent work (Gretton et al. (2009), Tillman et al. (2009), Zhang et al. (2011)) has… CONTINUE READING
Related Discussions
This paper has been referenced on Twitter 3 times. VIEW TWEETS

Citations

Publications citing this paper.
Showing 1-10 of 10 extracted citations

Estimating Causal Effects with Ancestral Graph Markov Models

Probabilistic Graphical Models • 2016
View 4 Excerpts
Highly Influenced

Subpopulation-Wise Conditional Correlation Modeling and Analysis

2017 IEEE International Conference on Big Knowledge (ICBK) • 2017
View 1 Excerpt

References

Publications referenced by this paper.
Showing 1-10 of 15 references

Applied Smoothing Techniques for Data Analysis

Technometrics • 1999
View 1 Excerpt
Highly Influenced

Partial association measures and an application to qualitative regression

J J Daudin
Biometrika • 1980
View 3 Excerpts
Highly Influenced

PC algorithm for nonparanormal graphical models

Journal of Machine Learning Research • 2013
View 2 Excerpts

A fast algorithm for evaluation of normalized Hermite functions. BIT Numerical Mathematics

B Bunck
A fast algorithm for evaluation of normalized Hermite functions. BIT Numerical Mathematics • 2009
View 1 Excerpt

Similar Papers

Loading similar papers…