Abstract
It is shown how the problem of estimating conditional Kendall's tau can be rewritten as a classification task. Conditional Kendall's tau is a conditional dependence parameter that is a characteristic of a given pair of random variables. The goal is to predict whether the pair is concordant (value of 1) or discordant (value of −1) conditionally on some covariates. The consistency and the asymptotic normality of a family of penalized approximate maximum likelihood estimators is proven, including the equivalent of the logit and probit regressions in our framework. Specific algorithms are detailed, adapting usual machine learning techniques, including nearest neighbors, decision trees, random forests and neural networks, to the setting of the estimation of conditional Kendall's tau. Finite sample properties of these estimators and their sensitivities to each component of the datagenerating process are assessed in a simulation study. Finally, all these estimators are applied to a dataset of European stock indices.
Original language  English 

Pages (fromto)  7094 
Number of pages  25 
Journal  Computational Statistics and Data Analysis 
Volume  135 
DOIs  
Publication status  Published  Jul 2019 
Externally published  Yes 
Keywords
 Classification task
 Conditional dependence measure
 Conditional Kendall's tau
 Machine learning
 Stock indices
Fingerprint
Dive into the research topics of 'A classification pointofview about conditional Kendall's tau'. Together they form a unique fingerprint.Datasets

CondCopulas: Estimation and Inference for Conditional Copula Models
Derumigny, A. F. F. (Creator), Fermanian, J. D. (Contributor), Min, A. (Contributor) & van der Spek, R. A. J. (Contributor), TU Delft  4TU.ResearchData, 2022
https://github.com/AlexisDerumigny/CondCopulas and one more link, https://CRAN.Rproject.org/package=CondCopulas (show fewer)
Dataset/Software: Software