A Bayesian Nonparametric Approach to Factor Analysis

Publikation: Working paperForskning

Standard

A Bayesian Nonparametric Approach to Factor Analysis. / Piatek, Rémi; Papaspiliopoulos, Omiros.

2018.

Publikation: Working paperForskning

Harvard

Piatek, R & Papaspiliopoulos, O 2018 'A Bayesian Nonparametric Approach to Factor Analysis'. <https://www.econ.ku.dk/piatek/pdf/BNPfactor.pdf>

APA

Piatek, R., & Papaspiliopoulos, O. (2018). A Bayesian Nonparametric Approach to Factor Analysis. https://www.econ.ku.dk/piatek/pdf/BNPfactor.pdf

Vancouver

Piatek R, Papaspiliopoulos O. A Bayesian Nonparametric Approach to Factor Analysis. 2018 jan.

Author

Piatek, Rémi ; Papaspiliopoulos, Omiros. / A Bayesian Nonparametric Approach to Factor Analysis. 2018.

Bibtex

@techreport{d3d7b82969834e1ca76dfdafb130ed64,
title = "A Bayesian Nonparametric Approach to Factor Analysis",
abstract = "This paper introduces a new approach for the inference of non-Gaussian factor models based on Bayesian nonparametric methods. It relaxes the usual normality assumption on the latent factors, widely used in practice, which is too restrictive in many settings. Our approach, on the contrary, does not impose any particular assumptions on the shape of the distribution of the factors, but still secures the basic requirements for the identification of the model. We design a new sampling scheme based on marginal data augmentation for the inference of mixtures of normals with location and scale restrictions. This approach is augmented by the use of a retrospective sampler, to allow for the inference of a constrained Dirichlet process mixture model for the distribution of the latent factors. We carry out a simulation study to illustrate the methodology and demonstrate its benefits. Our sampler is very efficient in recovering the distribution of the factors, and only generates models that fulfill the identification requirements. A real data example illustrates the applicability of the approach.",
author = "R{\'e}mi Piatek and Omiros Papaspiliopoulos",
year = "2018",
month = jan,
language = "English",
type = "WorkingPaper",

}

RIS

TY - UNPB

T1 - A Bayesian Nonparametric Approach to Factor Analysis

AU - Piatek, Rémi

AU - Papaspiliopoulos, Omiros

PY - 2018/1

Y1 - 2018/1

N2 - This paper introduces a new approach for the inference of non-Gaussian factor models based on Bayesian nonparametric methods. It relaxes the usual normality assumption on the latent factors, widely used in practice, which is too restrictive in many settings. Our approach, on the contrary, does not impose any particular assumptions on the shape of the distribution of the factors, but still secures the basic requirements for the identification of the model. We design a new sampling scheme based on marginal data augmentation for the inference of mixtures of normals with location and scale restrictions. This approach is augmented by the use of a retrospective sampler, to allow for the inference of a constrained Dirichlet process mixture model for the distribution of the latent factors. We carry out a simulation study to illustrate the methodology and demonstrate its benefits. Our sampler is very efficient in recovering the distribution of the factors, and only generates models that fulfill the identification requirements. A real data example illustrates the applicability of the approach.

AB - This paper introduces a new approach for the inference of non-Gaussian factor models based on Bayesian nonparametric methods. It relaxes the usual normality assumption on the latent factors, widely used in practice, which is too restrictive in many settings. Our approach, on the contrary, does not impose any particular assumptions on the shape of the distribution of the factors, but still secures the basic requirements for the identification of the model. We design a new sampling scheme based on marginal data augmentation for the inference of mixtures of normals with location and scale restrictions. This approach is augmented by the use of a retrospective sampler, to allow for the inference of a constrained Dirichlet process mixture model for the distribution of the latent factors. We carry out a simulation study to illustrate the methodology and demonstrate its benefits. Our sampler is very efficient in recovering the distribution of the factors, and only generates models that fulfill the identification requirements. A real data example illustrates the applicability of the approach.

M3 - Working paper

BT - A Bayesian Nonparametric Approach to Factor Analysis

ER -

ID: 168865371