A Projection-based Conditional Dependence Measure with Applications to High-dimensional Undirected Graphical Models.
By Jianqing Fan, Yang Feng, Lucy Xia. 2015
34 citations recorded by the source corpus.
Measuring conditional dependence is an important topic in econometrics with broad applications including graphical models. Under a factor model setting, a new conditional dependence measure based on projection is proposed. The corresponding conditional independence test is developed with the asymptotic null distribution unveiled where the number of factors could be high-dimensional. It is also shown that the new test has control over the asymptotic type I error and can be calculated efficiently. A generic method for building dependency graphs without Gaussian assumption using the new test is elaborated. We show the superiority of the new method, implemented in the R package pgraph, through simulation and real data studies.
Subfields: Artificial Intelligence, Statistics and Probability
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Rajen Dinesh Shah
- Richard J. Samworth
- Thomas B. Berrett
- Jan Peters
- Sergey Levine
- Pieter Abbeel
- Bernhard Schölkopf
- J. Urgen Schmidhuber
In Statistics and Probability
- Rajen Dinesh Shah
- Thomas B. Berrett
- Jan Peters
- Richard J. Samworth
- Peter Buhlmann
- Bernhard Schölkopf
- Mathias Drton
- Nicolai Meinshausen
Citations in the corpus
- The hardness of conditional independence testing and the generalised covariance measure (2018, background): “[21] propose to use in the case of additive noise models.”
- Nonparametric independence testing via mutual information (2017, background): “Existing tests of conditional independence include the proposals of Su and White (2008), Zhang et al. (2011) and Fan, Feng and Xia (2017).”