Statistical Methods and Bayesian Inference

On Motivating the Choice of Link Function in Binomial Regression Through A Latent Abundance Process

This paper introduces a framework for choosing link functions in binomial regression based on latent abundance processes, proposing a new generalized complementary log-log link for overdispersed counts.

Imputing relative abundances via Gaussian random fields

The paper introduces a Gaussian random field model for imputing missing data in longitudinal microbiome studies, achieving comparable accuracy to baselines while modeling temporal and multivariate correlations.

Imputing missing counts for waterbird species: A comparison of methods for zero-inflated biodiversity monitoring data

The paper compares imputation methods for zero-inflated biodiversity data, finding that statistical models best handle zero counts and uncertainty, while contrast-based approaches match them in predictions.

Characterizing Individual and Population-Level Heterogeneity in Infant BMI Trajectories Using Flexible and Functional Modeling Approaches

The paper models infant BMI trajectories using flexible methods to identify individual and population-level patterns, finding sex differences and associations with maternal gestational diabetes.

Modeling Patient-Reported Pain Trajectories with Frequent Minimum and Maximum Scores

The paper introduces a Bayesian beta-binomial mixed-effects model to analyze zero- and one-inflated patient-reported pain scores, showing improved fit and accuracy compared to existing methods.

A method for analyzing physiological data with multiple non-independent observations

The paper introduces a method to analyze physiological data with multiple non-independent observations, ensuring each subject is represented once and without distribution assumptions.

Bayesian Factor Analysis for Binary and Ordinal Phenotypes with Missingness

This paper introduces FABOr, a Bayesian framework for matrix factorization of binary and ordinal phenotypes, showing improved imputation accuracy on real-world data.

FEMA-Long: Modeling unstructured covariances for discovery of time-dependent effects in large-scale longitudinal datasets

FEMA-Long enables efficient modeling of time-dependent effects in longitudinal data, discovering genetic variants with time-dependent effects on infant growth traits.

The Hypothesis Race Model for evaluation of research findings

The paper introduces the Hypothesis Race Model as a Bayesian approach to improve the evaluation of research findings by addressing limitations of Null Hypothesis Significance Testing.

Regression-based Modeling of Spearman's Rho for Longitudinal Metabolomics and Mental Wellness in Breast Cancer Patients

The paper develops a regression framework to analyze longitudinal metabolomics data and mental wellness in breast cancer patients, identifying metabolites associated with mental health changes over time.

Improving Statistical Rigor in Animal Aging Research by Addressing Clustering and Nesting Effects: Illustration with the National Institute on Aging's Intervention Testing Program Data

This paper demonstrates how accounting for clustering and nesting effects in lifespan studies improves statistical rigor and reduces Type I error rates using data from the National Institute on Aging's Intervention Testing Program.

Between- and Within-subject Variability in Actigraphy Data: A Case Study on Sleep Patterns in Schizophrenia

The paper introduces a mixed-effects location-scale model to analyze variability in longitudinal actigraphy data, showing that modeling both mean and variability improves understanding of sleep patterns in schizophrenia.

Joint Modeling of Longitudinal Biomarker and Survival Outcomes with the Presence of Competing Risk in Nested Case-Control Studies with Application to the TEDDY Microbiome Dataset

The paper introduces JM-NCC, a joint modeling framework for nested case-control studies with competing risks, applied to the TEDDY microbiome dataset to analyze biomarker and survival outcomes.

Hierarchical Bayesian modeling of multi-region brain cell count data

The paper applies hierarchical Bayesian modeling to multi-region brain cell count data, showing improved inference over traditional methods in under-sampled datasets.

Statistical Methods for Binary Outcomes Adjusting for Outcome Dependent Sampling in Longitudinal Studies with Nonignorable Dropout

The paper develops statistical methods to adjust for nonignorable dropout in longitudinal studies with outcome-dependent sampling, using mixture models and the BUILD R package.

The Art of Not Knowing: Accommodating Structured Missingness in Biomedical Research

This paper addresses structured missingness in biomedical data by introducing a hierarchical imputation framework and advocating for alternative evaluation metrics.

Missing data approaches for longitudinal neuroimaging research: Examples from the Adolescent Brain and Cognitive Development (ABCD) Study

The paper discusses methods for handling missing data in longitudinal neuroimaging datasets, using the ABCD Study as an example, and advocates for advanced statistical techniques over conventional approaches.

hmde: Hierarchical Methods for Differential Equations

The paper introduces an R package for hierarchical Bayesian longitudinal modeling of growth data, providing tools for model fitting and case studies.

Practical Bayesian Inference in Neuroscience: Or How I Learned To Stop Worrying and Embrace the Distribution

The paper provides a practical guide to Bayesian inference in neuroscience, demonstrating its application through regression and ANOVA models with rodent auditory data and offering an open-source toolkit.

Adaptive predictor-set linear model: an imputation-free method for linear regression prediction on datasets with missing values

This paper introduces aps-lm, a linear regression method that predicts outcomes from incomplete datasets without imputation, showing improved accuracy and reduced bias compared to imputation strategies.

Non-Gaussian Normative Modelling With Hierarchical Bayesian Regression

This paper extends hierarchical Bayesian regression to model non-Gaussian data with skewness and kurtosis, showing improved performance on neuroimaging datasets.

LIMBARE: an Advanced Linear Mixed-effects Breakpoint Analysis with Robust Estimation Method with Applications to Longitudinal Ophthalmic Studies

LIMBARE improves breakpoint estimation in longitudinal ophthalmic studies by using robust linear mixed-effects modeling, outperforming other methods in simulations and real data.

lmeEEG: Mass linear mixed-effects modeling of EEG data with crossed random effects

The paper introduces lmeEEG, a method for applying linear mixed-effects models to EEG data with crossed random effects, showing improved performance and computational efficiency.

Statistical power and false positive rates for interdependent outcomes are strongly influenced by test type: Implications for behavioral neuroscience

The paper evaluates statistical methods for interdependent outcomes in behavioral neuroscience, finding high false positives with linear models and limited power with alternatives.

Increasing information gain in animal research by improving statistical model accuracy

The paper improves statistical model accuracy in animal research to increase information gain per animal, showing benefits for small sample sizes.

A real data-driven simulation strategy to select an imputation method for mixed-type trait data

The paper evaluates imputation methods for mixed-type trait data using a real data-driven simulation strategy, finding that random forest with phylogenetic information performs best.

A monotone single index model for missing-at-random longitudinal proportion data

The paper develops a semi-parametric Beta regression model for longitudinal proportion data with missing-at-random responses, using a monotone single index transform and Bayesian inference.

Systematic Bayesian posterior analysis guided by Kullback-Leibler divergence facilitates hypothesis formation

This paper introduces a KL divergence-based Bayesian framework to systematically rank parameters for hypothesis formation, demonstrating its effectiveness in two computational models.

Accessible Analysis of Longitudinal Data with Linear Mixed Effects Models

The paper re-analyzes longitudinal mouse growth data using linear mixed effects models and compares them to ANOVA, showing inconsistencies and the importance of modeling correlated data correctly.

An empirically-driven guide on using Bayes Factors for M/EEG decoding

This paper provides a guide on using Bayes Factors for M/EEG decoding results, examining their impact on statistical assessment of neural data.