Journal
JOURNAL OF MACHINE LEARNING RESEARCH
Volume 22, Issue -, Pages -Publisher
MICROTOME PUBL
Keywords
Conditional independence testing; nonparametric testing; partial copula; conditional distribution function; quantile regression
Funding
- VILLUM FONDEN [13358]
Ask authors/readers for more resources
This paper develops a nonparametric test for conditional independence by combining the partial copula with a quantile regression based method for estimating the nonparametric residuals. The resulting test is demonstrated to be sound under complicated data generating distributions and competitive to other state-of-the-art conditional independence tests, with superior power in cases with conditional variance heterogeneity of X and Y given Z.
The partial copula provides a method for describing the dependence between two random variables X and Y conditional on a third random vector Z in terms of nonparametric residuals U-1 and U-2. This paper develops a nonparametric test for conditional independence by combining the partial copula with a quantile regression based method for estimating the nonparametric residuals. We consider a test statistic based on generalized correlation between U-1 and U-2 and derive its large sample properties under consistency assumptions on the quantile regression procedure. We demonstrate through a simulation study that the resulting test is sound under complicated data generating distributions. Moreover, in the examples considered the test is competitive to other state-of-the-art conditional independence tests in terms of level and power, and it has superior power in cases with conditional variance heterogeneity of X and Y given Z.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available