Measuring dependence powerfully and equitably
From MaRDI portal
Abstract: Given a high-dimensional data set we often wish to find the strongest relationships within it. A common strategy is to evaluate a measure of dependence on every variable pair and retain the highest-scoring pairs for follow-up. This strategy works well if the statistic used is equitable [Reshef et al. 2015a], i.e., if, for some measure of noise, it assigns similar scores to equally noisy relationships regardless of relationship type (e.g., linear, exponential, periodic). In this paper, we introduce and characterize a population measure of dependence called MIC*. We show three ways that MIC* can be viewed: as the population value of MIC, a highly equitable statistic from [Reshef et al. 2011], as a canonical "smoothing" of mutual information, and as the supremum of an infinite sequence defined in terms of optimal one-dimensional partitions of the marginals of the joint distribution. Based on this theory, we introduce an efficient approach for computing MIC* from the density of a pair of random variables, and we define a new consistent estimator MICe for MIC* that is efficiently computable. In contrast, there is no known polynomial-time algorithm for computing the original equitable statistic MIC. We show through simulations that MICe has better bias-variance properties than MIC. We then introduce and prove the consistency of a second statistic, TICe, that is a trivial side-product of the computation of MICe and whose goal is powerful independence testing rather than equitability. We show in simulations that MICe and TICe have good equitability and power against independence respectively. The analyses here complement a more in-depth empirical evaluation of several leading measures of dependence [Reshef et al. 2015b] that shows state-of-the-art performance for MICe and TICe.
Recommendations
- An empirical study of the maximal and total information coefficients and leading measures of dependence
- A robust-equitable measure for feature ranking and selection
- Detecting novel associations in large data sets
- Equitability, mutual information, and the maximal information coefficient
- A measure of general functional dependence between two continuous variables
Cited in
(24)- Detecting novel associations in large data sets
- Robust dependence measure for detecting associations in large data set
- The randomized information coefficient: assessing dependencies in noisy data
- An empirical study of the maximal and total information coefficients and leading measures of dependence
- Four simple axioms of dependence measures
- A method for extracting nonlinear structure based on measures of dependence
- Using tours to visually investigate properties of new projection pursuit indexes with application to problems in physics
- Equitability, interval estimation, and statistical power
- Validation of association
- Equitability, mutual information, and the maximal information coefficient
- scientific article; zbMATH DE number 4040934 (Why is no real title available?)
- A robust-equitable measure for feature ranking and selection
- Generalized R-squared for detecting dependence
- Distribution-Free Consistent Independence Tests via Center-Outward Ranks and Signs
- The Hellinger Correlation
- Topological biclustering ARTMAP for identifying within bicluster relationships
- Multivariate Rank-Based Distribution-Free Nonparametric Testing Using Measure Transportation
- Adapted Chatterjee correlation coefficient
- On marginal feature attributions of tree-based models
- Application of hybrid feature selection methods to improve turbulence models using machine learning
- Transport dependency: optimal transport based dependency measures
- Using Rejection Sampling Probability of Acceptance as a Measure of Independence
- Detecting dependence structure: visualization and inference
- Estimating scale-invariant directed dependence of bivariate distributions
This page was built for publication: Measuring dependence powerfully and equitably
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2953635)
