MétaCan
Menu
Back to cohort
Record W6989101572

Accounting for heterogeneity in the dependence mechanism of longitudinal data

2022· dissertation· en· W6989101572 on OpenAlexaboutno aff

Bibliographic record

VenueMspace (University of Manitoba) · 2022
Typedissertation
Languageen
FieldEconomics, Econometrics and Finance
TopicSpatial and Panel Data Analysis
Canadian institutionsnot available
Fundersnot available
KeywordsCholesky decompositionCovarianceCopula (linguistics)Covariance matrixUnivariateMissing dataLongitudinal dataBayesian probabilityRandom effects model
DOInot available

Abstract

fetched live from OpenAlex

Longitudinal data occur frequently in practice where measurements are collected from subjects over time with an aim to understand the dependence mechanisms among these measurements. A major challenge in longitudinal data analysis is the presence of a complex dependence structure due to both between and within individual heterogeneity. This thesis develops new statistical methodologies that incorporate potential heterogeneity in the dependence structure in various longitudinal data problems. In the first part, we introduce a D-vine copula-based heterogeneous dependence model which provides a flexible representation of time-heterogeneous dependence in univariate longitudinal data with a continuous outcome. The proposed model allows for time adjustment in the dependence structure of unequally spaced and potentially unbalanced longitudinal data. We show that the proposed approach offers flexibility over its time-homogeneous counterparts as well as allows for parsimonious model specifications at the tree or vine level for a given D-vine structure. The performances of the time-heterogeneous D-vine copula models are evaluated through simulation studies and by real data from the Manitoba Follow-up Study. In the second part, we propose an approach to incorporate potential heterogeneity in the random effects covariance matrix in longitudinal data with missing responses and mismeasured covariates. The proposed approach uses a modified Cholesky decomposition and allows the random effects covariance matrix to depend on covariates. This decomposition provides an unconstrained and statistically meaningful reparameterization of the random effect covariance matrix which can be modeled without the concern of positive definiteness of the resulting estimators. The performance of the proposed approach is evaluated through simulation studies and is demonstrated using longitudinal data from Framingham Heart Study. In the last part, we review two major statistical models for longitudinal functional data that are spatially correlated and propose a computationally efficient modeling approach by incorporating a spatio-temporal dependence structure in the error process. Numerical experiments are conducted to compare these models and to investigate the impact of ignoring spatial correlation on prediction performance. We discuss the limitations of these models and outline future directions to develop flexible models that can incorporate potential heterogeneity in the dependence structure of spatial longitudinal data.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.028
metaresearch head score (Gemma)0.077
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Theoretical or conceptual · Consensus signal: none
GenreCandidate signal: Methods · Consensus signal: Methods
Teacher disagreement score0.028
Threshold uncertainty score0.150

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0280.077
Meta-epidemiology (narrow)0.0010.001
Meta-epidemiology (broad)0.0020.003
Bibliometrics0.0020.003
Science and technology studies0.0010.001
Scholarly communication0.0030.003
Open science0.0030.003
Research integrity0.0020.003
Insufficient payload (model declined to judge)0.0020.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.074
GPT teacher head0.245
Teacher spread0.172 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designTheoretical or conceptual
Domainnot available
GenreMethods

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2022
Admission routes1
Has abstractyes

Explore more

Same venueMspace (University of Manitoba)Same topicSpatial and Panel Data AnalysisFrench-language works237,207