Nonlinear Regime-Switching State-Space (RSSS) Models

Sy-Miin Chow, Guangjian Zhang

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

Nonlinear dynamic factor analysis models extend standard linear dynamic factor analysis models by allowing time series processes to be nonlinear at the latent level (e.g., involving interaction between two latent processes). In practice, it is often of interest to identify the phases-namely, latent "regimes" or classes-during which a system is characterized by distinctly different dynamics. We propose a new class of models, termed nonlinear regime-switching state-space (RSSS) models, which subsumes regime-switching nonlinear dynamic factor analysis models as a special case. In nonlinear RSSS models, the change processes within regimes, represented using a state-space model, are allowed to be nonlinear. An estimation procedure obtained by combining the extended Kalman filter and the Kim filter is proposed as a way to estimate nonlinear RSSS models. We illustrate the utility of nonlinear RSSS models by fitting a nonlinear dynamic factor analysis model with regime-specific cross-regression parameters to a set of experience sampling affect data. The parallels between nonlinear RSSS models and other well-known discrete change models in the literature are discussed briefly.

Original languageEnglish (US)
Pages (from-to)740-768
Number of pages29
JournalPsychometrika
Volume78
Issue number4
DOIs
StatePublished - Oct 1 2013

Fingerprint

Regime-switching Model
Space Simulation
State-space Model
Statistical Factor Analysis
Nonlinear Dynamics
Factor Analysis
Dynamic Analysis
Factor analysis
Latent Process
Regime Switching
Model
Kalman Filter
Standard Model
Time series
Regression
Filter
Extended Kalman filters
Interaction

All Science Journal Classification (ASJC) codes

  • Psychology(all)
  • Applied Mathematics

Cite this

Chow, Sy-Miin ; Zhang, Guangjian. / Nonlinear Regime-Switching State-Space (RSSS) Models. In: Psychometrika. 2013 ; Vol. 78, No. 4. pp. 740-768.
@article{4a0dd4a58b7f4925a9d852b3bf1efa76,
title = "Nonlinear Regime-Switching State-Space (RSSS) Models",
abstract = "Nonlinear dynamic factor analysis models extend standard linear dynamic factor analysis models by allowing time series processes to be nonlinear at the latent level (e.g., involving interaction between two latent processes). In practice, it is often of interest to identify the phases-namely, latent {"}regimes{"} or classes-during which a system is characterized by distinctly different dynamics. We propose a new class of models, termed nonlinear regime-switching state-space (RSSS) models, which subsumes regime-switching nonlinear dynamic factor analysis models as a special case. In nonlinear RSSS models, the change processes within regimes, represented using a state-space model, are allowed to be nonlinear. An estimation procedure obtained by combining the extended Kalman filter and the Kim filter is proposed as a way to estimate nonlinear RSSS models. We illustrate the utility of nonlinear RSSS models by fitting a nonlinear dynamic factor analysis model with regime-specific cross-regression parameters to a set of experience sampling affect data. The parallels between nonlinear RSSS models and other well-known discrete change models in the literature are discussed briefly.",
author = "Sy-Miin Chow and Guangjian Zhang",
year = "2013",
month = "10",
day = "1",
doi = "10.1007/s11336-013-9330-8",
language = "English (US)",
volume = "78",
pages = "740--768",
journal = "Psychometrika",
issn = "0033-3123",
publisher = "Springer New York",
number = "4",

}

Nonlinear Regime-Switching State-Space (RSSS) Models. / Chow, Sy-Miin; Zhang, Guangjian.

In: Psychometrika, Vol. 78, No. 4, 01.10.2013, p. 740-768.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Nonlinear Regime-Switching State-Space (RSSS) Models

AU - Chow, Sy-Miin

AU - Zhang, Guangjian

PY - 2013/10/1

Y1 - 2013/10/1

N2 - Nonlinear dynamic factor analysis models extend standard linear dynamic factor analysis models by allowing time series processes to be nonlinear at the latent level (e.g., involving interaction between two latent processes). In practice, it is often of interest to identify the phases-namely, latent "regimes" or classes-during which a system is characterized by distinctly different dynamics. We propose a new class of models, termed nonlinear regime-switching state-space (RSSS) models, which subsumes regime-switching nonlinear dynamic factor analysis models as a special case. In nonlinear RSSS models, the change processes within regimes, represented using a state-space model, are allowed to be nonlinear. An estimation procedure obtained by combining the extended Kalman filter and the Kim filter is proposed as a way to estimate nonlinear RSSS models. We illustrate the utility of nonlinear RSSS models by fitting a nonlinear dynamic factor analysis model with regime-specific cross-regression parameters to a set of experience sampling affect data. The parallels between nonlinear RSSS models and other well-known discrete change models in the literature are discussed briefly.

AB - Nonlinear dynamic factor analysis models extend standard linear dynamic factor analysis models by allowing time series processes to be nonlinear at the latent level (e.g., involving interaction between two latent processes). In practice, it is often of interest to identify the phases-namely, latent "regimes" or classes-during which a system is characterized by distinctly different dynamics. We propose a new class of models, termed nonlinear regime-switching state-space (RSSS) models, which subsumes regime-switching nonlinear dynamic factor analysis models as a special case. In nonlinear RSSS models, the change processes within regimes, represented using a state-space model, are allowed to be nonlinear. An estimation procedure obtained by combining the extended Kalman filter and the Kim filter is proposed as a way to estimate nonlinear RSSS models. We illustrate the utility of nonlinear RSSS models by fitting a nonlinear dynamic factor analysis model with regime-specific cross-regression parameters to a set of experience sampling affect data. The parallels between nonlinear RSSS models and other well-known discrete change models in the literature are discussed briefly.

UR - http://www.scopus.com/inward/record.url?scp=84884981699&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84884981699&partnerID=8YFLogxK

U2 - 10.1007/s11336-013-9330-8

DO - 10.1007/s11336-013-9330-8

M3 - Article

C2 - 24092487

AN - SCOPUS:84884981699

VL - 78

SP - 740

EP - 768

JO - Psychometrika

JF - Psychometrika

SN - 0033-3123

IS - 4

ER -