Filters








7,637 Hits in 5.3 sec

Causality on Longitudinal Data: Stable Specification Search in Constrained Structural Equation Modeling [article]

Ridho Rahmadi, Perry Groot, Marieke HC van Rijn, Jan AJG van den Brand, Marianne Heins, Hans Knoop, Tom Heskes (the Alzheimer's Disease Neuroimaging Initiatives, the MASTERPLAN Study Group, the OPTIMISTIC Consortium)
2017 arXiv   pre-print
A typical problem in causal modeling is the instability of model structure learning, i.e., small changes in finite data can result in completely different optimal models.  ...  We represent causal relationships using structural equation models. Models are scored along two objectives: the model fit and the model complexity.  ...  Stable specification search for longitudinal data Stable Specification Search for Longitudinal data (S3L) is an extension of S3C.  ... 
arXiv:1605.06838v3 fatcat:kegaehaetbajnhbxxqcjgulbyq

Causality on longitudinal data: Stable specification search in constrained structural equation modeling

Ridho Rahmadi, Perry Groot, Marieke HC van Rijn, Jan AJG van den Brand, Marianne Heins, Hans Knoop, Tom Heskes
2017 Statistical Methods in Medical Research  
We model causal models using structural equation models. Models are scored along two objectives: the model fit and the model complexity.  ...  Developing causal models from observational longitudinal studies is an important, ubiquitous problem in many disciplines.  ...  Stable Specification Search Our method, stable specification search [13] , can be divided into two phases (see Figure 2 ). The first phase is search, performing exploratory search over SEM models.  ... 
doi:10.1177/0962280217713347 pmid:28657454 fatcat:g32a6pjhzbbitevqxbiviagj7y

Causality on cross-sectional data: Stable specification search in constrained structural equation modeling

Ridho Rahmadi, Perry Groot, Marianne Heins, Hans Knoop, Tom Heskes
2017 Applied Soft Computing  
The present work introduces a new hypothesis-free score-based causal discovery algorithm, called stable specification search, that is robust for finite samples based on recent advances in stability selection  ...  Structure search is performed over Structural Equation Models. Our approach uses exploratory search but allows incorporation of prior background knowledge.  ...  Structure search is performed over Structural Equation Models (SEM), which is the most widely used language for causal discovery in various scientific disciplines.  ... 
doi:10.1016/j.asoc.2016.10.003 fatcat:ymkinn4se5crpphx67i3a7gdxy

The stablespec package for causal discovery on cross-sectional and longitudinal data in R

Ridho Rahmadi, Perry Groot, Tom Heskes
2018 Neurocomputing  
The method aims at causal discovery on both cross-sectional and longitudinal data through stable specification search in constrained structural equation models. © 2017 Elsevier B.V.  ...  Stable specification search is a novel causal discovery method based on [5] , for cross-sectional data (S3C), and [6] , for longitudinal data (S3L).  ... 
doi:10.1016/j.neucom.2017.10.064 fatcat:3rqsmaaoznalnknglb3uqd7sqm

Semiparametric Estimation of the Impacts of Longitudinal Interventions on Adolescent Obesity using Targeted Maximum-Likelihood: Accessible Estimation with the ltmle Package

Anna L. Decker, Alan Hubbard, Catherine M. Crespi, Edmund Y.W. Seto, May C. Wang
2014 Journal of Causal Inference  
An alternative parameter to estimate is one motivated by the causal inference literature, which can be interpreted as the mean change in the outcome under interventions to set the exposure of interest.  ...  Our analysis demonstrates that sophisticated, optimal semiparametric estimation of longitudinal treatment-specific means via ltmle provides an incredibly powerful, yet easy-to-use tool, removing impediments  ...  In the longitudinal settings, G-computation is an identifiability result derived from the sequential randomization assumption implied by a causal graph or the nonparametric structural equation (NPSEM)  ... 
doi:10.1515/jci-2013-0025 pmid:26046009 pmcid:PMC4452010 fatcat:lrvfyfnltvfapdjjmeoycmatpm

Learning Subject-Specific Directed Acyclic Graphs With Mixed Effects Structural Equation Models From Observational Data

Xiang Li, Shanghong Xie, Peter McColgan, Sarah J Tabrizi, Rachael I Scahill, Donglin Zeng, Yuanjia Wang
2018 Frontiers in Genetics  
We propose a new mixed-effects structural equation model (mSEM) framework to estimate subject-specific DAGs, where we represent joint distribution of random variables in the DAG as a set of structural  ...  In addition, by pooling information across subject-specific DAGs, we can identify causal structure with a high probability and estimate subject-specific networks with a high precision.  ...  Leavitt and the Track, PREDICT Investigators for their contribution to collect TRACK-HD and PREDICT-HD data, and NINDS dbGap data repository (accession number phs000222.v3).  ... 
doi:10.3389/fgene.2018.00430 pmid:30333854 pmcid:PMC6176748 fatcat:7t7nnchawfbpnggef4jjiiy44y

Causal Analysis of Self-tracked Time Series Data Using a Counterfactual Framework for N-of-1 Trials

Eric Daza
2018 Methods of Information in Medicine  
the challenges we faced in conducting N1OS causal discovery.Causal analysis of an individual's time series data can be facilitated by an N1RT counterfactual framework.  ...  We then apply the framework and methods to search for estimable and interpretable APTEs using six years of the author's self-tracked weight and exercise data, and report both the preliminary findings and  ...  We define a data-generating process (DGP) to be such a time-constrained function (e.g., the univariate structural equations in White and Lu, 2010 [30] ), and call the structural equation expression of  ... 
doi:10.3414/me16-02-0044 pmid:29621835 pmcid:PMC6087468 fatcat:yboc7hntebbv5hcfrhs4npeof4

Panel data and models of change: A comparison of first difference and conventional two-wave models

Jeffrey K Liker, Sue Augustyniak, Greg J Duncan
1985 Social Science Research  
The greatest potential contribution of longitudinal data is that they permit empirical analyses of dynamic aspects of behavior.  ...  Panel data on individuals, families, and their life conditions are becoming increasingly available and open new possibilities for analysis. But they also present new challenges and problems.  ...  Given these theoretical and data limitations, major advances from the use of longitudinal data for the estimation of truly dynamic models are apt to be slow in coming.  ... 
doi:10.1016/0049-089x(85)90013-4 fatcat:5bpkxutgijdltfkqzxy2vxvzgm

Network Data [article]

Bryan S. Graham
2019 arXiv   pre-print
I emphasize (i) dyadic regression analysis incorporating unobserved agent-specific heterogeneity and supporting causal inference, (ii) techniques for estimating, and conducting inference on, summary network  ...  Many economic activities are embedded in networks: sets of agents and the (often) rivalrous relationships connecting them to one another.  ...  The second exploits the exponential family structure of the model and conditions on a sufficient statistic for A 0 . Both estimates have antecedents in the literature on panel data.  ... 
arXiv:1912.06346v1 fatcat:viy67tqtbrhvvoifldgjeqt3nu

The Gaussian Graphical Model in Cross-sectional and Time-series Data [article]

Sacha Epskamp, Lourens J. Waldorp, René Mõttus, Denny Borsboom
2018 arXiv   pre-print
The GGM shows which variables predict one-another, allows for sparse modeling of covariance structures, and may highlight potential causal relationships between observed variables.  ...  When analyzing data from multiple subjects, a GGM can also be formed on the covariance structure of stationary means---the between-subjects network.  ...  Acknowledgements We would like to thank Laura Bringmann, Noémi Schuurman, Oisín Ryan and Ellen Hamaker for helpful tips and invigorating discussions, and Katharina Jorgensen for valuable comments on earlier  ... 
arXiv:1609.04156v6 fatcat:sdbfaxapx5ed3ox6swzk4rdra4

At the Frontiers of Modeling Intensive Longitudinal Data: Dynamic Structural Equation Models for the Affective Measurements from the COGITO Study

E. L. Hamaker, T. Asparouhov, A. Brose, F. Schmiedek, B. Muthén
2018 Multivariate Behavioral Research  
Here we use dynamic multilevel modeling, as it is incorporated in the new dynamic structural equation modeling (DSEM) toolbox in Mplus, to analyze the affective data from the COGITO study.  ...  KEYWORDS Dynamic structural equation modeling (DSEM); intensive longitudinal data; dynamic multilevel modeling; multilevel time series analysis ABSTRACT With the growing popularity of intensive longitudinal  ...  Such structural missing data are quite common in intensive longitudinal data.  ... 
doi:10.1080/00273171.2018.1446819 pmid:29624092 fatcat:s3zw6mudqjcflngub3mildabuu

The Gaussian Graphical Model in Cross-Sectional and Time-Series Data

Sacha Epskamp, Lourens J. Waldorp, René Mõttus, Denny Borsboom
2018 Multivariate Behavioral Research  
The GGM shows which variables predict one-another, allows for sparse modeling of covariance structures, and may highlight potential causal relationships between observed variables.  ...  When analyzing data from multiple subjects, a GGM can also be formed on the covariance structure of stationary means-the between-subjects network.  ...  An earlier version of this paper has been adapted as a chapter in the dissertation of the main author (Epskamp, 2017b) .  ... 
doi:10.1080/00273171.2018.1454823 pmid:29658809 fatcat:23jaxodovzd2hnaxsrchchkoeu

Integrating molecular, histopathological, neuroimaging and clinical neuroscience data with NeuroPM-box

Yasser Iturria-Medina, Félix Carbonell, Atousa Assadi, Quadri Adewale, Ahmed F Khan, Tobias R Baumeister, Lazaro Sanchez-Rodriguez
2021 Communications Biology  
Using advanced analytical modeling for molecular, histopathological, brain-imaging and/or clinical evaluations, this framework has multiple applications, validated here with synthetic (N > 2900), in-vivo  ...  brain atrophy), and (iv) biologically-defined patient stratification based on disease heterogeneity and/or therapeutic needs.  ...  This project was also undertaken thanks in part to the following funding awards to Y.  ... 
doi:10.1038/s42003-021-02133-x pmid:34021244 fatcat:wlncznrxbzatxmsapuy22446fa

A Nonlinear Dynamic Model Applied to Data with Two Times of Measurement

Anton Grip, Lars R. Bergman
2016 Journal for Person-Oriented Research  
A good model of developmental phenomena should not only "explain" the data in the sense that the model is not falsified by the data but it should also be built on the basic theoretical assumptions about  ...  It is pointed out in the article that it might be possible to use nonlinear dynamic system modeling also in contexts where data are available only from a few measurement occasions.  ...  For instance, standard methods for measuring fit within the structural equation modeling tradition have limitations in that many fit measures are based on the similarity between the predicted and actual  ... 
doi:10.17505/jpor.2016.06 fatcat:huiffyxezbdlbkwbbyehedfnxy

Longitudinal Data Analysis with Structural Equations

Jesús Rosel, Ian Plewis
2008 Methodology: European Journal of Research Methods for the Behavioral and Social Sciences  
In this paper we review different structural equation models for the analysis of longitudinal data: (a) univariate models of observable variables, (b) multivariate models of observable variables, (c) models  ...  single level and multilevel measurement, and (i) other advances in SEM of longitudinal data (latent growth curve model, latent difference score, etc.).  ...  Longitudinal data analysis with structural equations Since Jöreskog (1969) , Keesling (1972) and Wiley (1973) first developed the statistical model of structural equations, it has become one of the  ... 
doi:10.1027/1614-2241.4.1.37 fatcat:qvduqyciszfolbacxjguu2k37m
« Previous Showing results 1 — 15 out of 7,637 results