Loglinear model selection and human mobility

Dobra, Adrian; Mohammadi, Abdolreza

doi:10.1214/18-aoas1164

Cited by 16 publications

(23 citation statements)

References 84 publications

(125 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this paper we make use of a Bayesian framework for solving the structural learning problem that is suitable for the analysis of hyper-sparse contingency tables with p = 48 variables. This framework [70] determines graphical loglinear models that are a special type of hierarchical loglinear models [8, 9]. A graphical model for a random vector X = ( X 1 , X 2 , …, X p ) is specified by an undirected graph G = ( V , E ) where V = {1, …, p } are vertices or nodes, and E ⊂ V × V are edges or links [9].…”

Section: Methodsmentioning

confidence: 99%

A method for statistical analysis of repeated residential movements to link human mobility and HIV acquisition

et al. 2019

Self Cite

View full text Add to dashboard Cite

We propose a method for analyzing repeated residential movements based on graphical loglinear models. This method allows an explicit representation of residential presence and absence patterns from several areas without defining mobility measures. We make use of our method to analyze data from one of the most comprehensive demographic surveillance sites in Africa that is characterized by high adult HIV prevalence, high levels of poverty and unemployment and frequent residential changes. Between 2004 and 2016, residential changes were recorded for 8,857 men over 35,500.01 person-years, and for 12,158 women over 57,945.35 person-years. These individuals were HIV negative at baseline. Over the study duration, there were a total of 806 HIV seroconversions in men, and 2,458 HIV seroconversions in women. Our method indicates that establishing a residence outside the rural study area is a strong predictor of HIV seroconversion in men (OR = 2.003, 95% CI = [1.718,2.332]), but not in women. Residing inside the rural study area in a single or in multiple locations is a less significant risk factor for HIV acquisition in both men and women compared to moving outside the rural study area.

show abstract

Section: Methodsmentioning

confidence: 99%

A method for statistical analysis of repeated residential movements to link human mobility and HIV acquisition

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…We do not apply thinning. The Bayesian copula method is implemented using the R package, BDGraph Wit, 2015, 2017;Dobra and Mohammadi, 2017;Mohammadi and Wit, 2018) using the option "gcgm". Posterior graph selection is done using Bayesian model averaging, the default option in the BDGraph package, in which it selects the graph with links for which their estimated posterior probabilities are greater than 0.5.…”

Section: Simulation Resultsmentioning

confidence: 99%

Bayesian Inference in Nonparanormal Graphical Models

Mulgrave¹,

Ghosal²

2020

Bayesian Anal.

View full text Add to dashboard Cite

A nonparanormal graphical model is a semiparametric generalization of a Gaussian graphical model for continuous variables in which it is assumed that the variables follow a Gaussian graphical model only after some unknown smooth monotone transformations. We consider a Bayesian approach in the nonparanormal graphical model in which we put priors on the unknown transformations through random series based on B-splines. We use a regression formulation to construct the likelihood through the Cholesky decomposition on the underlying precision matrix of the transformed variables and put shrinkage priors on the regression coefficients. We apply a plug-in variational Bayesian algorithm for learning the sparse precision matrix and compare the performance to a posterior Gibbs sampling scheme in a simulation study. We finally apply the proposed methods to a real data set.

show abstract

“…Selection of hierarchical loglinear models has been widely discussed in the statistical literature [21,19,1,57]. More recent approaches that work well for high-dimensional sparse contingency tables involve Bayesian Markov chain Monte Carlo (MCMC) algorithms [39,40,41,13,51,14,16,15,17].…”

Section: Approachmentioning

confidence: 99%

Identifying mediating variables with graphical models: an application to the study of causal pathways in people living with HIV

Dobra

Buhikire

Voss

2019

Journal of Applied Statistics

Self Cite

View full text Add to dashboard Cite

We empirically demonstrate that graphical models can be a valuable tool in the identification of mediating variables in causal pathways. We make use of graphical models to elucidate the causal pathway through which the treatment influences the levels of fatigue and weakness in people living with HIV (PLHIV) based on a secondary analysis of a categorical dataset collected in a behavioral clinical trial: is weakness a mediator for the treatment and fatigue, or is fatigue a mediator for the treatment and weakness? Causal mediation analysis could not offer any definite answers to these questions.

show abstract

Loglinear model selection and human mobility

Cited by 16 publications

References 84 publications

A method for statistical analysis of repeated residential movements to link human mobility and HIV acquisition

A method for statistical analysis of repeated residential movements to link human mobility and HIV acquisition

Bayesian Inference in Nonparanormal Graphical Models

Identifying mediating variables with graphical models: an application to the study of causal pathways in people living with HIV

Contact Info

Product

Resources

About