A model selection method based on maximal information coefficient of residuals

DOI10.1016/S0252-9602(14)60031-XMaRDI QIDQ2514923zbMATH OpenFDO

Authors Qiuheng Tan, Hangjin Jiang, Yiming Ding

Publication date 11 February 2015

Published in Acta Mathematica Scientia. Series B. (English Edition) (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/S0252-9602(14)60031-X

zbMATH Keywords

model selectionresidualmaximal information coefficientpartial maximal information coefficient

Mathematics Subject Classification ID

Statistical aspects of information-theoretic topics (62B10) Applications of statistics (62P99)

Recommendations

• Resolution dependence of the maximal information coefficient for noiseless relationship
• Regression Model Selection—A Residual Likelihood Approach
• Variable selection in regression using maximal correlation and distance correlation
• Residual information criterion for single-index model selections
• Model selection with data-oriented penalty

Cited in

(4)

• Non-stationarity measurement based on law of iterated logarithm
• Modelling financial time series based on heavy-tailed market microstructure models with scale mixtures of normal distributions
• Robust dependence measure for detecting associations in large data set
• A novel approach of dependence measure for complex signals

This page was built for publication: A model selection method based on maximal information coefficient of residuals

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2514923)