PARTIAL CORRELATION AND CONDITIONAL CORRELATION AS MEASURES OF CONDITIONAL INDEPENDENCE

By Kunihiro Baba, R. Shibata, M. Sibuya. 2004

492 citations recorded by the source corpus.

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

In Statistics and Probability

Citations in the corpus

Kernel Mean Embedding of Distributions: A Review and Beyonds (2016, methodology): “There exist numerous tests for conditional dependency such as partial correlation tests (Baba et al. 2004), conditional densities based tests (Su and White 2008), and permutation-based tests (Tsamardinos and Borboudakis 2010).”
50 Years of Test (Un)fairness: Lessons for Machine Learning (2018, background): “uivalent to sufficiency in the special case where A, R and Y have a multivariate Gaussian distribution. This is because for this special case the partial correlation ρAY.X = 0 is equivalent to A ⊥Y|R [3]. In general though, we cannot assume even a one way implication, since A ⊥Y|R does not imply ρAY”
Critical Assessment of Methods of Protein Structure Prediction (CASP) – Progress and New directions in Round XI (2016, methodology): “This kind of knock-on effect is in fact a well-known problem in statistics, when trying to deduce causation,17 and in statistical physics, between sets of interacting particles, for example in Ising models.18 The pre-existing methods for treating this problem have now been adapted for the contact…”
A cautionary note on robust covariance plug-in methods (2014, background): “In particular, if the regression ofy onx is linear, then conditional independence implies a zero partial correlation, see Theorem 1 in Baba et al. (2004).”
Selection of Vine Copulas (2013, background): “Since partial correlations are equal to conditional correlations for elliptical distributions (see [5]), this strategy is particularly appropriate if all pair copula families are elliptical.”
Communication-Avoiding Optimization Methods for Distributed Massive-Scale Sparse Inverse Covariance Estimation (2017, background): “Under the assumption that the underlying data-generating distribution is multivariate normal, the partial correlation graph is precisely the conditional independence graph associated with the variables [27, 26, 7].”
Partial Distance Correlation with Methods for Dissimilarities (2013, unknown): “According to Baba, Shibata and Sibuya [1], if (X, Y, Z) are trivariate normal, then pcor (X, Y ; Z) = 0 if and only if X and Y are conditionally independent given Z .”
Inverse statistical problems: from the inverse Ising problem to data science (2017, background): “This approach is also linked to the concept of partial correlations in statistics [9, 119].”
A Scalable Conditional Independence Test for Nonlinear, Non-Gaussian Data (2014, unknown): “ (The term "conditional correlation" is as in Baba et al., 2004.”
Big data for cyber physical systems in industry 4.0: a survey (2019, methodology): “Popular methods for detecting confounding factors include the Cochran-Mantel-Haenszel method (Cochran 1954), logistic regression model (Li et al. 2014), and partial correlation (Baba, Shibata, and Sibuya 2004).”