TY - JOUR
T1 - A Square-Root Second-Order Extended Kalman Filtering Approach for Estimating Smoothly Time-Varying Parameters
AU - Fisher, Zachary F.
AU - Chow, Sy Miin
AU - Molenaar, Peter C.M.
AU - Fredrickson, Barbara L.
AU - Pipiras, Vladas
AU - Gates, Kathleen M.
N1 - Funding Information:
Funding : The first author was supported by the Population Science Training Program (T32 HD007168) and general support of the Carolina Population Center (P2C HD050924). Vladas Pipiras’s research was partially supported by the grant NSF DMS 1712966.Sy-Miin Chow’s research was supported by NIH Intensive Longitudinal Health Behavior Cooperative Agreement Program U24AA027684, National Science Foundation grant IGE-1806874. Barbara L. Fredrickson’s research was supported by the National Cancer Institute (R01CA170128) of the U.S. National Institutes of Health (NIH).
Publisher Copyright:
© 2020 Taylor & Francis Group, LLC.
PY - 2022
Y1 - 2022
N2 - Researchers collecting intensive longitudinal data (ILD) are increasingly looking to model psychological processes, such as emotional dynamics, that organize and adapt across time in complex and meaningful ways. This is also the case for researchers looking to characterize the impact of an intervention on individual behavior. To be useful, statistical models must be capable of characterizing these processes as complex, time-dependent phenomenon, otherwise only a fraction of the system dynamics will be recovered. In this paper we introduce a Square-Root Second-Order Extended Kalman Filtering approach for estimating smoothly time-varying parameters. This approach is capable of handling dynamic factor models where the relations between variables underlying the processes of interest change in a manner that may be difficult to specify in advance. We examine the performance of our approach in a Monte Carlo simulation and show the proposed algorithm accurately recovers the unobserved states in the case of a bivariate dynamic factor model with time-varying dynamics and treatment effects. Furthermore, we illustrate the utility of our approach in characterizing the time-varying effect of a meditation intervention on day-to-day emotional experiences.
AB - Researchers collecting intensive longitudinal data (ILD) are increasingly looking to model psychological processes, such as emotional dynamics, that organize and adapt across time in complex and meaningful ways. This is also the case for researchers looking to characterize the impact of an intervention on individual behavior. To be useful, statistical models must be capable of characterizing these processes as complex, time-dependent phenomenon, otherwise only a fraction of the system dynamics will be recovered. In this paper we introduce a Square-Root Second-Order Extended Kalman Filtering approach for estimating smoothly time-varying parameters. This approach is capable of handling dynamic factor models where the relations between variables underlying the processes of interest change in a manner that may be difficult to specify in advance. We examine the performance of our approach in a Monte Carlo simulation and show the proposed algorithm accurately recovers the unobserved states in the case of a bivariate dynamic factor model with time-varying dynamics and treatment effects. Furthermore, we illustrate the utility of our approach in characterizing the time-varying effect of a meditation intervention on day-to-day emotional experiences.
UR - http://www.scopus.com/inward/record.url?scp=85092258194&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85092258194&partnerID=8YFLogxK
U2 - 10.1080/00273171.2020.1815513
DO - 10.1080/00273171.2020.1815513
M3 - Article
C2 - 33025834
AN - SCOPUS:85092258194
VL - 57
SP - 134
EP - 152
JO - Multivariate Behavioral Research
JF - Multivariate Behavioral Research
SN - 0027-3171
IS - 1
ER -