A Comparison of Two-Stage Approaches for Fitting Nonlinear Ordinary Differential Equation Models with Mixed Effects

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

Several approaches exist for estimating the derivatives of observed data for model exploration purposes, including functional data analysis (FDA; Ramsay & Silverman, 2005), generalized local linear approximation (GLLA; Boker, Deboeck, Edler, & Peel, 2010), and generalized orthogonal local derivative approximation (GOLD; Deboeck, 2010). These derivative estimation procedures can be used in a two-stage process to fit mixed effects ordinary differential equation (ODE) models. While the performance and utility of these routines for estimating linear ODEs have been established, they have not yet been evaluated in the context of nonlinear ODEs with mixed effects. We compared properties of the GLLA and GOLD to an FDA-based two-stage approach denoted herein as functional ordinary differential equation with mixed effects (FODEmixed) in a Monte Carlo (MC) study using a nonlinear coupled oscillators model with mixed effects. Simulation results showed that overall, the FODEmixed outperformed both the GLLA and GOLD across all the embedding dimensions considered, but a novel use of a fourth-order GLLA approach combined with very high embedding dimensions yielded estimation results that almost paralleled those from the FODEmixed. We discuss the strengths and limitations of each approach and demonstrate how output from each stage of FODEmixed may be used to inform empirical modeling of young children’s self-regulation.

Original languageEnglish (US)
Pages (from-to)154-184
Number of pages31
JournalMultivariate Behavioral Research
Volume51
Issue number2-3
DOIs
StatePublished - May 3 2016

Fingerprint

Mixed Effects
Nonlinear Ordinary Differential Equations
Ordinary differential equation
Functional Differential Equations
Model
Derivative Estimation
Nonlinear ODE
Functional Data Analysis
Derivative
Local Approximation
Nonlinear Oscillator
Linear Approximation
Coupled Oscillators
Monte Carlo Study
Ordinary Differential Equations
Fourth Order
Output
Approximation
Modeling
Demonstrate

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Experimental and Cognitive Psychology
  • Arts and Humanities (miscellaneous)

Cite this

@article{29d9fc0649464379b3d2dba3ad15557a,
title = "A Comparison of Two-Stage Approaches for Fitting Nonlinear Ordinary Differential Equation Models with Mixed Effects",
abstract = "Several approaches exist for estimating the derivatives of observed data for model exploration purposes, including functional data analysis (FDA; Ramsay & Silverman, 2005), generalized local linear approximation (GLLA; Boker, Deboeck, Edler, & Peel, 2010), and generalized orthogonal local derivative approximation (GOLD; Deboeck, 2010). These derivative estimation procedures can be used in a two-stage process to fit mixed effects ordinary differential equation (ODE) models. While the performance and utility of these routines for estimating linear ODEs have been established, they have not yet been evaluated in the context of nonlinear ODEs with mixed effects. We compared properties of the GLLA and GOLD to an FDA-based two-stage approach denoted herein as functional ordinary differential equation with mixed effects (FODEmixed) in a Monte Carlo (MC) study using a nonlinear coupled oscillators model with mixed effects. Simulation results showed that overall, the FODEmixed outperformed both the GLLA and GOLD across all the embedding dimensions considered, but a novel use of a fourth-order GLLA approach combined with very high embedding dimensions yielded estimation results that almost paralleled those from the FODEmixed. We discuss the strengths and limitations of each approach and demonstrate how output from each stage of FODEmixed may be used to inform empirical modeling of young children’s self-regulation.",
author = "Chow, {Sy Miin} and Bendez{\'u}, {Jason J.} and Cole, {Pamela M.} and Nilam Ram",
year = "2016",
month = "5",
day = "3",
doi = "10.1080/00273171.2015.1123138",
language = "English (US)",
volume = "51",
pages = "154--184",
journal = "Multivariate Behavioral Research",
issn = "0027-3171",
publisher = "Psychology Press Ltd",
number = "2-3",

}

TY - JOUR

T1 - A Comparison of Two-Stage Approaches for Fitting Nonlinear Ordinary Differential Equation Models with Mixed Effects

AU - Chow, Sy Miin

AU - Bendezú, Jason J.

AU - Cole, Pamela M.

AU - Ram, Nilam

PY - 2016/5/3

Y1 - 2016/5/3

N2 - Several approaches exist for estimating the derivatives of observed data for model exploration purposes, including functional data analysis (FDA; Ramsay & Silverman, 2005), generalized local linear approximation (GLLA; Boker, Deboeck, Edler, & Peel, 2010), and generalized orthogonal local derivative approximation (GOLD; Deboeck, 2010). These derivative estimation procedures can be used in a two-stage process to fit mixed effects ordinary differential equation (ODE) models. While the performance and utility of these routines for estimating linear ODEs have been established, they have not yet been evaluated in the context of nonlinear ODEs with mixed effects. We compared properties of the GLLA and GOLD to an FDA-based two-stage approach denoted herein as functional ordinary differential equation with mixed effects (FODEmixed) in a Monte Carlo (MC) study using a nonlinear coupled oscillators model with mixed effects. Simulation results showed that overall, the FODEmixed outperformed both the GLLA and GOLD across all the embedding dimensions considered, but a novel use of a fourth-order GLLA approach combined with very high embedding dimensions yielded estimation results that almost paralleled those from the FODEmixed. We discuss the strengths and limitations of each approach and demonstrate how output from each stage of FODEmixed may be used to inform empirical modeling of young children’s self-regulation.

AB - Several approaches exist for estimating the derivatives of observed data for model exploration purposes, including functional data analysis (FDA; Ramsay & Silverman, 2005), generalized local linear approximation (GLLA; Boker, Deboeck, Edler, & Peel, 2010), and generalized orthogonal local derivative approximation (GOLD; Deboeck, 2010). These derivative estimation procedures can be used in a two-stage process to fit mixed effects ordinary differential equation (ODE) models. While the performance and utility of these routines for estimating linear ODEs have been established, they have not yet been evaluated in the context of nonlinear ODEs with mixed effects. We compared properties of the GLLA and GOLD to an FDA-based two-stage approach denoted herein as functional ordinary differential equation with mixed effects (FODEmixed) in a Monte Carlo (MC) study using a nonlinear coupled oscillators model with mixed effects. Simulation results showed that overall, the FODEmixed outperformed both the GLLA and GOLD across all the embedding dimensions considered, but a novel use of a fourth-order GLLA approach combined with very high embedding dimensions yielded estimation results that almost paralleled those from the FODEmixed. We discuss the strengths and limitations of each approach and demonstrate how output from each stage of FODEmixed may be used to inform empirical modeling of young children’s self-regulation.

UR - http://www.scopus.com/inward/record.url?scp=84978128558&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84978128558&partnerID=8YFLogxK

U2 - 10.1080/00273171.2015.1123138

DO - 10.1080/00273171.2015.1123138

M3 - Article

C2 - 27391255

AN - SCOPUS:84978128558

VL - 51

SP - 154

EP - 184

JO - Multivariate Behavioral Research

JF - Multivariate Behavioral Research

SN - 0027-3171

IS - 2-3

ER -