MCMC for Imbalanced Categorical Data

James E. Johndrow, Aaron Smith, Natesh Pillai and David B. Dunson

Journal of the American Statistical Association, 2019, vol. 114, issue 527, 1394-1403

Abstract: Many modern applications collect highly imbalanced categorical data, with some categories relatively rare. Bayesian hierarchical models combat data sparsity by borrowing information, while also quantifying uncertainty. However, posterior computation presents a fundamental barrier to routine use; a single class of algorithms does not work well in all settings and practitioners waste time trying different types of Markov chain Monte Carlo (MCMC) approaches. This article was motivated by an application to quantitative advertising in which we encountered extremely poor computational performance for data augmentation MCMC algorithms but obtained excellent performance for adaptive Metropolis. To obtain a deeper understanding of this behavior, we derive theoretical results on the computational complexity of commonly used data augmentation algorithms and the Random Walk Metropolis algorithm for highly imbalanced binary data. In this regime, our results show computational complexity of Metropolis is logarithmic in sample size, while data augmentation is polynomial in sample size. The root cause of this poor performance of data augmentation is a discrepancy between the rates at which the target density and MCMC step sizes concentrate. Our methods also show that MCMC algorithms that exhibit a similar discrepancy will fail in large samples—a result with substantial practical impact. Supplementary materials for this article are available online.

Date: 2019
References: Add references at CitEc
Citations: View citations in EconPapers (7)

Downloads: (external link)
http://hdl.handle.net/10.1080/01621459.2018.1505626 (text/html)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:taf:jnlasa:v:114:y:2019:i:527:p:1394-1403

Ordering information: This journal article can be ordered from
http://www.tandfonline.com/pricing/journal/UASA20

DOI: 10.1080/01621459.2018.1505626

Access Statistics for this article

Journal of the American Statistical Association is currently edited by Xuming He, Jun Liu, Joseph Ibrahim and Alyson Wilson

More articles in Journal of the American Statistical Association from Taylor & Francis Journals
Bibliographic data for series maintained by Chris Longhurst ().