Skip to main content

Mixture modeling on related samples by ψ-stick breaking and kernel perturbation

Publication ,  Journal Article
Soriano, J; Ma, L
Published in: Bayesian Analysis
January 1, 2019

There has been great interest recently in applying nonparametric kernel mixtures in a hierarchical manner to model multiple related data samples jointly. In such settings several data features are commonly present: (i) the related samples often share some, if not all, of the mixture components but with differing weights, (ii) only some, not all, of the mixture components vary across the samples, and (iii) often the shared mixture components across samples are not aligned perfectly in terms of their kernel parameters such as the location and spread in Gaussian kernels, but rather display small misalignments either due to systematic cross-sample difference or more often due to uncontrolled, extraneous causes. Properly incorporating these features in mixture modeling will enhance the efficiency of inference, whereas ignoring them not only reduces efficiency but can jeopardize the validity of the inference due to issues such as confounding. We propose to use two techniques for incorporating these features in modeling related data samples using kernel mixtures. The first technique, called ψ-stick breaking, is a joint generative process for the mixing weights through the breaking of both a stick shared by all the samples for the components that do not vary in size across samples and an idiosyncratic stick for each sample for those components that do vary in size. The second technique is to imbue random perturbation into the kernels, thereby accounting for cross-sample misalignment. These techniques can be used either separately or together in both parametric and nonparametric kernel mixtures. We derive efficient Bayesian inference recipes based on Markov Chain Monte Carlo (MCMC) sampling for models featuring these techniques, and illustrate their work through both simulated data and a real flow cytometry data set in prediction/estimation and testing multi-sample differences.

Duke Scholars

Published In

Bayesian Analysis

DOI

EISSN

1931-6690

ISSN

1936-0975

Publication Date

January 1, 2019

Volume

14

Issue

1

Start / End Page

161 / 180

Related Subject Headings

  • Statistics & Probability
  • 4905 Statistics
  • 0104 Statistics
 

Citation

APA
Chicago
ICMJE
MLA
NLM
Soriano, J., & Ma, L. (2019). Mixture modeling on related samples by ψ-stick breaking and kernel perturbation. Bayesian Analysis, 14(1), 161–180. https://doi.org/10.1214/18-BA1106
Soriano, J., and L. Ma. “Mixture modeling on related samples by ψ-stick breaking and kernel perturbation.” Bayesian Analysis 14, no. 1 (January 1, 2019): 161–80. https://doi.org/10.1214/18-BA1106.
Soriano J, Ma L. Mixture modeling on related samples by ψ-stick breaking and kernel perturbation. Bayesian Analysis. 2019 Jan 1;14(1):161–80.
Soriano, J., and L. Ma. “Mixture modeling on related samples by ψ-stick breaking and kernel perturbation.” Bayesian Analysis, vol. 14, no. 1, Jan. 2019, pp. 161–80. Scopus, doi:10.1214/18-BA1106.
Soriano J, Ma L. Mixture modeling on related samples by ψ-stick breaking and kernel perturbation. Bayesian Analysis. 2019 Jan 1;14(1):161–180.

Published In

Bayesian Analysis

DOI

EISSN

1931-6690

ISSN

1936-0975

Publication Date

January 1, 2019

Volume

14

Issue

1

Start / End Page

161 / 180

Related Subject Headings

  • Statistics & Probability
  • 4905 Statistics
  • 0104 Statistics