Gaussian Process Vine Copulas for Multivariate Dependence

02/16/2013
by   David Lopez-Paz, et al.
0

Copulas allow to learn marginal distributions separately from the multivariate dependence structure (copula) that links them together into a density function. Vine factorizations ease the learning of high-dimensional copulas by constructing a hierarchy of conditional bivariate copulas. However, to simplify inference, it is common to assume that each of these conditional bivariate copulas is independent from its conditioning variables. In this paper, we relax this assumption by discovering the latent functions that specify the shape of a conditional copula given its conditioning variables We learn these functions by following a Bayesian approach based on sparse Gaussian processes with expectation propagation for scalable, approximate inference. Experiments on real-world datasets show that, when modeling all conditional dependencies, we obtain better estimates of the underlying copula of the data.

READ FULL TEXT
research
07/01/2013

Gaussian Process Conditional Copulas with Applications to Financial Time Series

The estimation of dependencies between multiple variables is a central p...
research
03/04/2021

Approximate Bayesian Conditional Copulas

Copula models are flexible tools to represent complex structures of depe...
research
11/08/2020

Pathwise Conditioning of Gaussian Processes

As Gaussian processes are integrated into increasingly complex problem s...
research
10/30/2018

Gaussian Process Conditional Density Estimation

Conditional Density Estimation (CDE) models deal with estimating conditi...
research
06/14/2022

Minimum information dependence modeling

We propose a method of dependence modeling for a broad class of multivar...
research
08/21/2020

Testing for equality between conditional copulas given discretized conditioning events

Several procedures have been recently proposed to test the simplifying a...
research
02/01/2014

Markov Blanket Ranking using Kernel-based Conditional Dependence Measures

Developing feature selection algorithms that move beyond a pure correlat...

Please sign up or login with your details

Forgot password? Click here to reset