Article
Mathematics, Interdisciplinary Applications
Holly P. O'Rourke, Kimberly L. Fine, Kevin J. Grimm, David P. MacKinnon
Summary: The precision of time metric has a significant impact on the estimation, interpretation, and accuracy of parameter estimates in bivariate latent change score models. Results show that using a coarse time metric can lead to biased parameter estimates, larger standard errors, and larger variances and covariances, particularly for unique coupling parameter estimates in BLCS models. Illustrative examples further highlight the importance of considering time metric precision in data analysis.
MULTIVARIATE BEHAVIORAL RESEARCH
(2022)
Article
Psychology, Multidisciplinary
Eduardo Estrada, Silvia A. Bunge, Emilio Ferrer
Summary: This article proposes an approach to identify and control for cohort effects in accelerated longitudinal designs. A simulation study is used to examine conditions related to sampling design, effect size of cohort effect, parameters affected by cohort effects, and modeling approach. The findings indicate that the proposed extension is effective for detecting and controlling for cohort effects equivalent to those documented in the literature. Models that included parameters to account for cohort effects correctly estimated them as null in most scenarios. Based on these results, the authors recommend using models that include parameters to account for cohort effects, especially LCS models in continuous time, when examining accelerated longitudinal designs data.
PSYCHOLOGICAL METHODS
(2023)
Article
Psychology, Multidisciplinary
Pablo F. Cancer, Eduardo Estrada, Mar J. F. Ollero, Emilio Ferrer
Summary: This paper explores the meaning and interpretation of parameters in Latent Change Score models (LCS), emphasizing the relationship between parameters and trajectory shape, as well as the specific assumptions about mechanisms of change involved in different specifications of the LCS model. Additionally, the paper presents an interactive Shiny App for users to explore the effects of different parameter values on predicted trajectories, along with fully explained code to estimate relevant specifications of the LCS model using R-packages lavaan and OpenMx.
FRONTIERS IN PSYCHOLOGY
(2021)
Article
Chemistry, Multidisciplinary
Sung-Jung Hsiao, Wen-Tsai Sung
Summary: This paper uses the approach of the differential model to effectively improve the analysis of particle swarm optimization, proposing a differential evolution PSO model. By adjusting parameters and using different transformation methods, the performance of the PSO algorithm is optimized.
APPLIED SCIENCES-BASEL
(2022)
Article
Psychology, Applied
James Garrett Matusik, John R. Hollenbeck, Rebecca L. Mitchell
Summary: This article discusses the challenges of studying change in organizational science and the potential solutions provided by latent change score models originating from developmental psychology. It emphasizes the need for cautious use of these models to avoid unwarranted rejection by the field.
ORGANIZATIONAL RESEARCH METHODS
(2021)
Article
Psychology, Clinical
Candice M. Monson, Philippe Shnaider, Anne C. Wagner, Rachel E. Liebman, Nicole D. Pukay-Martin, Meredith S. H. Landy, Sonya G. Wanklyn, Michael Suvak, Tae L. Hart, Naomi Koerner
Summary: This study examined the longitudinal associations between interpersonal relationship functioning and PTSD symptoms in recently traumatized individuals. The results showed that there were nonlinear declines in negative aspects of relationship functioning, but linear declines in positive aspects. The relationship between PTSD and relationship functioning differed for clinician- and self-reported PTSD.
PSYCHOLOGICAL MEDICINE
(2023)
Article
Automation & Control Systems
Weijing Tang, Jiaqi Ma, Qiaozhu Mei, Ji Zhu
Summary: This paper proposes a flexible model for survival analysis using neural networks and scalable optimization algorithms. It addresses the technical challenge of applying maximum likelihood estimation to censored data by recognizing it as a differential-equation constrained optimization problem. The model effectively evaluates the likelihood and gradients using efficient solvers and sensitivity analysis.
JOURNAL OF MACHINE LEARNING RESEARCH
(2022)
Article
Automation & Control Systems
Na-Na Jin, Shuai Liu, Huan-Shui Zhang
Summary: This paper investigates the linear optimal estimation problems of discrete-time and continuous-time systems with multiple state delays in measurements. The linear optimal estimation of state is obtained by direct calculation of optimal gain or analytical expression, without resorting to Riccati equations, saving computational cost. Infinite horizon case and recovery of Kalman filter are also studied with stability analysis.
INTERNATIONAL JOURNAL OF CONTROL AUTOMATION AND SYSTEMS
(2021)
Article
Automation & Control Systems
Qiang Xiao, Zhigang Zeng, Tingwen Huang, Frank L. Lewis
Summary: This paper investigates positivity and asymptotic stability for a class of timescale-type differential-difference equation with time-varying delay. Necessary and sufficient conditions for bounded and unbounded delay cases are obtained, and results are extended to corresponding continuous-time systems.
Article
Economics
Giuseppe Buccheri, Fulvio Corsi, Franco Flandoli, Giulia Livieri
Summary: This study examines the convergence of a class of discrete-time volatility models to a stochastic differential equation as the observation interval approaches zero. The form of the diffusion limit is influenced by factors such as the link function, the conditional second moment of the score, and the normalization of the score. These findings have significant implications for the estimation and filtering of diffusion processes.
JOURNAL OF ECONOMETRICS
(2021)
Article
Mathematics, Interdisciplinary Applications
Zareen A. Khan, Hijaz Ahmad
Summary: Discrete fractional calculus is utilized to interpret neural schemes with memory impacts. This study formulates discrete fractional nonlinear inequalities for numerical solutions of discrete fractional differential equations and discusses aspects such as boundedness, uniqueness, and continuous dependency. The methodology employed proves the leading consequences for solutions of discrete fractional difference equations.
FRACTALS-COMPLEX GEOMETRY PATTERNS AND SCALING IN NATURE AND SOCIETY
(2021)
Article
Mathematics, Applied
Alexandros Beskos, Dan Crisan, Ajay Jasra, Nikolas Kantas, Hamza Ruzayqat
Summary: The paper discusses parameter estimation for a class of continuous-time state space models, focusing on partially observed diffusion. Two particle filter based methodologies are considered to estimate the score function, with one using Euler discretization and standard particle smoothers, and the other based on a novel diffusion bridge construction. Multilevel methodologies are also explored to improve computational costs, with the parameter estimation method demonstrated through stochastic gradient approaches in numerical examples.
SIAM JOURNAL ON SCIENTIFIC COMPUTING
(2021)
Article
Mathematics, Interdisciplinary Applications
Yu-Yu Hsiao, Oi-Man Kwok, Mark H. C. Lai
Summary: The study compared the performance of LMS and RAPI methods on testing interaction effects, and discussed the effects of four reliability estimates. The results showed that the four reliability estimates performed comparably well in estimating interaction effects, and recommendations were provided on choosing different methods and reliability estimates based on sample size and population reliability conditions.
STRUCTURAL EQUATION MODELING-A MULTIDISCIPLINARY JOURNAL
(2021)
Article
Psychology, Multidisciplinary
Rita Cerutti, Andrea Fontana, Valerio Ghezzi, Francesca Menozzi, Valentina Spensieri, Renata Tambelli
Summary: Psychological distress is common among university students. This study found that a brief psychological intervention in an Italian university counseling service effectively reduced psychopathological symptoms in students who sought help.
CURRENT PSYCHOLOGY
(2022)
Article
Psychology, Developmental
Pascal R. Deboeck, David A. Cole, Kristopher J. Preacher, Rex Forehand, Bruce E. Compas
Summary: The study introduces continuous time models, highlighting that effective interventions may involve changes in how variables affect each other over time. The models can compare expected covariance matrices for different underlying processes, allowing for comparisons with a relatively small number of repeated observations.
INTERNATIONAL JOURNAL OF BEHAVIORAL DEVELOPMENT
(2021)