Continuous-discrete state-space modeling of panel data with nonlinear filter algorithms

Singer, Hermann

doi:10.1007/s10182-011-0172-3

Cited by 13 publications

(3 citation statements)

References 94 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although in the field of econometrics continuous time dynamic models have existed much longer than latent change score models (cf. Bergstrom, 1988), they are only slowly diffusing into psychological research (see Chow, Lu, Sherwood, & Zhu, 2016;Oravecz, Tuerlinckx, & Vandekerckhove, 2009, 2011Ou, Hunter, & Chow, 2017;Oud & Jansen, 2000;Oud & Singer, 2008;Singer, 2010Singer, , 2011Singer, , 2012Voelkle & Oud, 2013 for examples). From the latent change score formulation in Equation 3, it is only a small step to a continuous time model.…”

Section: Continuous Time Dynamic Modelsmentioning

confidence: 99%

The Role of Time in the Quest for Understanding Psychological Mechanisms

Voelkle

Gische

Driver

et al. 2018

Multivariate Behavioral Research

104

116

View full text Add to dashboard Cite

The lead-lag structure of multivariate time-ordered observations and the possibility to disentangle between-person (BP) from within-person (WP) sources of variance are major assets of longitudinal (panel) data. Hence, psychologists are making increasing use of such data, often with the intent to delineate the dynamic properties of psychological mechanisms, understood as a sequence of causal effects that govern psychological functioning. However, even with longitudinal data, psychological mechanisms are not easily identified. In this article, we show how an adequate representation of time may enhance the tenability of causal interpretations in the context of multivariate longitudinal data analysis. We anchor our considerations with an example that illustrates some of the main problems and questions faced by applied researchers and practitioners. We distinguish between static versus dynamic and discrete versus continuous time modeling approaches and discuss their advantages and disadvantages. We place particular emphasis on different ways of addressing BP differences and stress their dual role as potential confounds versus valuable sources of information for improving estimation and aiding causal inference. We conclude by outlining an approach that offers the potential of better integration of information on BP differences and WP changes in the search for causal mechanisms along with a discussion of current problems and limitations.

show abstract

Section: Continuous Time Dynamic Modelsmentioning

confidence: 99%

The Role of Time in the Quest for Understanding Psychological Mechanisms

Voelkle

Gische

Driver

et al. 2018

Multivariate Behavioral Research

104

116

View full text Add to dashboard Cite

show abstract

“…Despite the nonlinear connection between discrete and continuous time, the stochastic differential equation on which the procedure and ctsem are based, is linear. This differs from nonlinear SSMs that several authors in the behavioral and related sciences have turned attention to recently (e.g., Chow et al, 2016;King, Nguyen, & Ionides, 2016;Molenaar & Newell, 2003;Singer, 2011).…”

Section: Introductionmentioning

confidence: 80%

SEM Based CARMA Time Series Modeling for Arbitrary N

Oud

Voelkle

Driver

2017

Multivariate Behavioral Research

View full text Add to dashboard Cite

This article explains in detail the state space specification and estimation of first and higher-order autoregressive moving-average models in continuous time (CARMA) in an extended structural equation modeling (SEM) context for N = 1 as well as N > 1. To illustrate the approach, simulations will be presented in which a single panel model (T = 41 time points) is estimated for a sample of N = 1,000 individuals as well as for samples of N = 100 and N = 50 individuals, followed by estimating 100 separate models for each of the one-hundred N = 1 cases in the N = 100 sample. Furthermore, we will demonstrate how to test the difference between the full panel model and each N = 1 model by means of a subject-group-reproducibility test. Finally, the proposed analyses will be applied in an empirical example, in which the relationships between mood at work and mood at home are studied in a sample of N = 55 women. All analyses are carried out by ctsem, an R-package for continuous time modeling, interfacing to OpenMx.

show abstract

“…For a more comprehensive introduction to continuous time models in psychological research, see Voelkle et al (2012). For a more technical treatment, see Oud and Jansen (2000) and the work by Singer, (2011). For an overview of recent developments in the behavioural and related sciences, see van Montfort, Oud, and Voelkle (2018).…”

Section: A Quick Review Of Continuous Time Dynamic Modellingmentioning

confidence: 99%

Using Dynamic Panel Models to Study Age–Related Differences and Time–Related Changes in Personality

2019

View full text Add to dashboard Cite

Along with an increasing interest in the plasticity and role of personality across the adult lifespan comes the need for a diverse set of innovative statistical approaches to study it. With this paper, we set out to illustrate some of the possibilities and challenges in modelling age-related differences and time-related changes in personality psychology by means of dynamic panel models. To this end, we first distinguish between the study of age-related differences and timerelated changes and demonstrate how the treatment of age and time as either discrete or continuous variables implies important modelling choices. Second, we present a selection of four example cases that address the topic of age moderation in diverse matters and with different objectives. Based on our cross-tabulation of age and time as discrete and continuous variables, the first two example cases represent fairly well-established models (cases A and B), whereas the remaining cases are used to illustrate current developments in the field (cases C and D). We close the paper with some final remarks on current limitation and future research directions.

show abstract

Continuous-discrete state-space modeling of panel data with nonlinear filter algorithms

Abstract: Continuous-discrete state space models, Stochastic differential equations, Itô calculus, Sampling, Kalman filtering, Approximate nonlinear filtering, Structural equations modeling, Spatial models, Random fields, Stochastic partial differential equations,

Cited by 13 publications

References 94 publications

The Role of Time in the Quest for Understanding Psychological Mechanisms

The Role of Time in the Quest for Understanding Psychological Mechanisms

SEM Based CARMA Time Series Modeling for Arbitrary N

Using Dynamic Panel Models to Study Age–Related Differences and Time–Related Changes in Personality

Contact Info

Product

Resources

About