A monte carlo approach to nonnormal and nonlinear state–space modeling

Bradley P. Carlin; Nicholas G. Polson; David S. Stoffer

doi:10.1080/01621459.1992.10475231

A monte carlo approach to nonnormal and nonlinear state–space modeling

Bradley P. Carlin, Nicholas G. Polson, David S. Stoffer

Biostatistics

Research output: Contribution to journal › Article › peer-review

89 Scopus citations

Abstract

A solution to multivariate state-space modeling, forecasting, and smoothing is discussed. We allow for the possibilities of nonnormal errors and nonlinear functionals in the state equation, the observational equation, or both. An adaptive Monte Carlo integration technique known as the Gibbs sampler is proposed as a mechanism for implementing a conceptually and computationally simple solution in such a framework. The methodology is a general strategy for obtaining marginal posterior densities of coefficients in the model or of any of the unknown elements of the state space. Missing data problems (including the k-step ahead prediction problem) also are easily incorporated into this framework. We illustrate the broad applicability of our approach with two examples: a problem involving nonnormal error distributions in a linear model setting and a one-step ahead prediction problem in a situation where both the state and observational equations are nonlinear and involve unknown parameters.

Original language	English (US)
Pages (from-to)	493-500
Number of pages	8
Journal	Journal of the American Statistical Association
Volume	87
Issue number	418
DOIs	https://doi.org/10.1080/01621459.1992.10475231
State	Published - Jun 1992

Bibliographical note

Funding Information:
• Bradley P. Carlin is Assistant Professor, Division of Biostatistics, School of Public Health, University of Minnesota, Minneapolis, MN 55455. Nicholas G. Polson is Assistant Professor, Graduate School of Business, University of Chicago, Chicago, IL 60637. David S. Stoffer is Associate Professor, Department of Mathematics and Statistics, University of Pittsburgh, Pittsburgh, PA 15260. The work of Carlin was supported in part by National Science Foundation Grant DMS 88-05676; the work of Stoffer was supported in part by National Science Foundation Grant DMS 90-00522 and by a grant from the Centers for Disease Control through a cooperative agreement with the Association of Schools of Public Health. This research was done while all three authors were visitingthe Department of Statistics at Carnegie Mellon University.

Keywords

Forecasting
Gibbs sampler
Kalman filter
Smoothing

Publisher link

10.1080/01621459.1992.10475231

Find It

Find It at U of M Libraries

Cite this

@article{e66655fe2df64704aa3257266af8620e,

title = "A monte carlo approach to nonnormal and nonlinear state–space modeling",

abstract = "A solution to multivariate state-space modeling, forecasting, and smoothing is discussed. We allow for the possibilities of nonnormal errors and nonlinear functionals in the state equation, the observational equation, or both. An adaptive Monte Carlo integration technique known as the Gibbs sampler is proposed as a mechanism for implementing a conceptually and computationally simple solution in such a framework. The methodology is a general strategy for obtaining marginal posterior densities of coefficients in the model or of any of the unknown elements of the state space. Missing data problems (including the k-step ahead prediction problem) also are easily incorporated into this framework. We illustrate the broad applicability of our approach with two examples: a problem involving nonnormal error distributions in a linear model setting and a one-step ahead prediction problem in a situation where both the state and observational equations are nonlinear and involve unknown parameters.",

keywords = "Forecasting, Gibbs sampler, Kalman filter, Smoothing",

author = "Carlin, {Bradley P.} and Polson, {Nicholas G.} and Stoffer, {David S.}",

note = "Funding Information: • Bradley P. Carlin is Assistant Professor, Division of Biostatistics, School of Public Health, University of Minnesota, Minneapolis, MN 55455. Nicholas G. Polson is Assistant Professor, Graduate School of Business, University of Chicago, Chicago, IL 60637. David S. Stoffer is Associate Professor, Department of Mathematics and Statistics, University of Pittsburgh, Pittsburgh, PA 15260. The work of Carlin was supported in part by National Science Foundation Grant DMS 88-05676; the work of Stoffer was supported in part by National Science Foundation Grant DMS 90-00522 and by a grant from the Centers for Disease Control through a cooperative agreement with the Association of Schools of Public Health. This research was done while all three authors were visitingthe Department of Statistics at Carnegie Mellon University.",

year = "1992",

month = jun,

doi = "10.1080/01621459.1992.10475231",

language = "English (US)",

volume = "87",

pages = "493--500",

journal = "Journal of the American Statistical Association",

issn = "0162-1459",

publisher = "Taylor and Francis Ltd.",

number = "418",

}

TY - JOUR

T1 - A monte carlo approach to nonnormal and nonlinear state–space modeling

AU - Carlin, Bradley P.

AU - Polson, Nicholas G.

AU - Stoffer, David S.

N1 - Funding Information: • Bradley P. Carlin is Assistant Professor, Division of Biostatistics, School of Public Health, University of Minnesota, Minneapolis, MN 55455. Nicholas G. Polson is Assistant Professor, Graduate School of Business, University of Chicago, Chicago, IL 60637. David S. Stoffer is Associate Professor, Department of Mathematics and Statistics, University of Pittsburgh, Pittsburgh, PA 15260. The work of Carlin was supported in part by National Science Foundation Grant DMS 88-05676; the work of Stoffer was supported in part by National Science Foundation Grant DMS 90-00522 and by a grant from the Centers for Disease Control through a cooperative agreement with the Association of Schools of Public Health. This research was done while all three authors were visitingthe Department of Statistics at Carnegie Mellon University.

PY - 1992/6

Y1 - 1992/6

N2 - A solution to multivariate state-space modeling, forecasting, and smoothing is discussed. We allow for the possibilities of nonnormal errors and nonlinear functionals in the state equation, the observational equation, or both. An adaptive Monte Carlo integration technique known as the Gibbs sampler is proposed as a mechanism for implementing a conceptually and computationally simple solution in such a framework. The methodology is a general strategy for obtaining marginal posterior densities of coefficients in the model or of any of the unknown elements of the state space. Missing data problems (including the k-step ahead prediction problem) also are easily incorporated into this framework. We illustrate the broad applicability of our approach with two examples: a problem involving nonnormal error distributions in a linear model setting and a one-step ahead prediction problem in a situation where both the state and observational equations are nonlinear and involve unknown parameters.

AB - A solution to multivariate state-space modeling, forecasting, and smoothing is discussed. We allow for the possibilities of nonnormal errors and nonlinear functionals in the state equation, the observational equation, or both. An adaptive Monte Carlo integration technique known as the Gibbs sampler is proposed as a mechanism for implementing a conceptually and computationally simple solution in such a framework. The methodology is a general strategy for obtaining marginal posterior densities of coefficients in the model or of any of the unknown elements of the state space. Missing data problems (including the k-step ahead prediction problem) also are easily incorporated into this framework. We illustrate the broad applicability of our approach with two examples: a problem involving nonnormal error distributions in a linear model setting and a one-step ahead prediction problem in a situation where both the state and observational equations are nonlinear and involve unknown parameters.

KW - Forecasting

KW - Gibbs sampler

KW - Kalman filter

KW - Smoothing

UR - http://www.scopus.com/inward/record.url?scp=84950445183&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84950445183&partnerID=8YFLogxK

U2 - 10.1080/01621459.1992.10475231

DO - 10.1080/01621459.1992.10475231

M3 - Article

AN - SCOPUS:84950445183

SN - 0162-1459

VL - 87

SP - 493

EP - 500

JO - Journal of the American Statistical Association

JF - Journal of the American Statistical Association

IS - 418

ER -

A monte carlo approach to nonnormal and nonlinear state–space modeling

Abstract

Bibliographical note

Keywords

Publisher link

Find It

Other files and links

Fingerprint

Cite this