Variational Bayes estimation of discrete-margined copula models with application to time series

Rubén Loaiza-Maya, Michael Stanley Smith

Research output: Contribution to journalArticleResearchpeer-review

Abstract

We propose a new variational Bayes estimator for high-dimensional copulas with discrete, or a combination of discrete and continuous, margins. The method is based on a variational approximation to a tractable augmented posterior, and is faster than previous likelihood-based approaches. We use it to estimate drawable vine copulas for univariate and multivariate Markov ordinal and mixed time series. These have dimension rT, where T is the number of observations and r is the number of series, and are difficult to estimate using previous methods. The vine pair-copulas are carefully selected to allow for heteroskedasticity, which is a feature of most ordinal time series data. When combined with flexible margins, the resulting time series models also allow for other common features of ordinal data, such as zero inflation, multiple modes and under- or over-dispersion. Using six example series, we illustrate both the flexibility of the time series copula models, and the efficacy of the variational Bayes estimator for copulas of up to 792 dimensions and 60 parameters. This far exceeds the size and complexity of copula models for discrete data that can be estimated using previous methods. An Online Appendix and MATLAB code implementing the method are available as Supplementary Materials.
Original languageEnglish
Number of pages52
JournalJournal of Computational and Graphical Statistics
DOIs
Publication statusAccepted/In press - 14 Jan 2019
Externally publishedYes

Keywords

  • data augmentation
  • drawable vines
  • heteroskedasticity
  • multivariate ordinal and mixed time series
  • sparse variational approximation
  • stochastic gradient ascent

Cite this

@article{ad733af818424aa996590f5492a12b1e,
title = "Variational Bayes estimation of discrete-margined copula models with application to time series",
abstract = "We propose a new variational Bayes estimator for high-dimensional copulas with discrete, or a combination of discrete and continuous, margins. The method is based on a variational approximation to a tractable augmented posterior, and is faster than previous likelihood-based approaches. We use it to estimate drawable vine copulas for univariate and multivariate Markov ordinal and mixed time series. These have dimension rT, where T is the number of observations and r is the number of series, and are difficult to estimate using previous methods. The vine pair-copulas are carefully selected to allow for heteroskedasticity, which is a feature of most ordinal time series data. When combined with flexible margins, the resulting time series models also allow for other common features of ordinal data, such as zero inflation, multiple modes and under- or over-dispersion. Using six example series, we illustrate both the flexibility of the time series copula models, and the efficacy of the variational Bayes estimator for copulas of up to 792 dimensions and 60 parameters. This far exceeds the size and complexity of copula models for discrete data that can be estimated using previous methods. An Online Appendix and MATLAB code implementing the method are available as Supplementary Materials.",
keywords = "data augmentation, drawable vines, heteroskedasticity, multivariate ordinal and mixed time series, sparse variational approximation, stochastic gradient ascent",
author = "Rub{\'e}n Loaiza-Maya and Smith, {Michael Stanley}",
year = "2019",
month = "1",
day = "14",
doi = "10.1080/10618600.2018.1562936",
language = "English",
journal = "Journal of Computational and Graphical Statistics",
issn = "1061-8600",
publisher = "Taylor & Francis",

}

Variational Bayes estimation of discrete-margined copula models with application to time series. / Loaiza-Maya, Rubén; Smith, Michael Stanley.

In: Journal of Computational and Graphical Statistics, 14.01.2019.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Variational Bayes estimation of discrete-margined copula models with application to time series

AU - Loaiza-Maya, Rubén

AU - Smith, Michael Stanley

PY - 2019/1/14

Y1 - 2019/1/14

N2 - We propose a new variational Bayes estimator for high-dimensional copulas with discrete, or a combination of discrete and continuous, margins. The method is based on a variational approximation to a tractable augmented posterior, and is faster than previous likelihood-based approaches. We use it to estimate drawable vine copulas for univariate and multivariate Markov ordinal and mixed time series. These have dimension rT, where T is the number of observations and r is the number of series, and are difficult to estimate using previous methods. The vine pair-copulas are carefully selected to allow for heteroskedasticity, which is a feature of most ordinal time series data. When combined with flexible margins, the resulting time series models also allow for other common features of ordinal data, such as zero inflation, multiple modes and under- or over-dispersion. Using six example series, we illustrate both the flexibility of the time series copula models, and the efficacy of the variational Bayes estimator for copulas of up to 792 dimensions and 60 parameters. This far exceeds the size and complexity of copula models for discrete data that can be estimated using previous methods. An Online Appendix and MATLAB code implementing the method are available as Supplementary Materials.

AB - We propose a new variational Bayes estimator for high-dimensional copulas with discrete, or a combination of discrete and continuous, margins. The method is based on a variational approximation to a tractable augmented posterior, and is faster than previous likelihood-based approaches. We use it to estimate drawable vine copulas for univariate and multivariate Markov ordinal and mixed time series. These have dimension rT, where T is the number of observations and r is the number of series, and are difficult to estimate using previous methods. The vine pair-copulas are carefully selected to allow for heteroskedasticity, which is a feature of most ordinal time series data. When combined with flexible margins, the resulting time series models also allow for other common features of ordinal data, such as zero inflation, multiple modes and under- or over-dispersion. Using six example series, we illustrate both the flexibility of the time series copula models, and the efficacy of the variational Bayes estimator for copulas of up to 792 dimensions and 60 parameters. This far exceeds the size and complexity of copula models for discrete data that can be estimated using previous methods. An Online Appendix and MATLAB code implementing the method are available as Supplementary Materials.

KW - data augmentation

KW - drawable vines

KW - heteroskedasticity

KW - multivariate ordinal and mixed time series

KW - sparse variational approximation

KW - stochastic gradient ascent

U2 - 10.1080/10618600.2018.1562936

DO - 10.1080/10618600.2018.1562936

M3 - Article

JO - Journal of Computational and Graphical Statistics

JF - Journal of Computational and Graphical Statistics

SN - 1061-8600

ER -