Computational Statistical Methods for Social Network Models

Hunter, David R.; Krivitsky, Pavel N.; Schweinberger, Michael

doi:10.1080/10618600.2012.732921

Cited by 85 publications

(79 citation statements)

References 92 publications

(133 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…, we provide a data‐ and question‐driven approach to selecting the most suitable statistical tool. For further comparisons between statistical models of networks, and guidance to their usage, we refer readers to recent reviews in other subject areas (Hunter, Krivitsky & Schweinberger ; Leifeld & Cranmer ; Cranmer et al . ).…”

Section: Choosing a Modelmentioning

confidence: 99%

The application of statistical network models in disease research

et al. 2017

View full text Add to dashboard Cite

Summary1. Host social structure is fundamental to how infections spread and persist, and so the statistical modelling of static and dynamic social networks provides an invaluable tool to parameterise realistic epidemiological models. 2. We present a practical guide to the application of network modelling frameworks for hypothesis testing related to social interactions and epidemiology, illustrating some approaches with worked examples using data from a population of wild European badgers Meles meles naturally infected with bovine tuberculosis. 3. Different empirical network datasets generate particular statistical issues related to non-independence and sampling constraints. We therefore discuss the strengths and weaknesses of modelling approaches for different types of network data and for answering different questions relating to disease transmission. 4. We argue that statistical modelling frameworks designed specifically for network analysis offer great potential in directly relating network structure to infection. They have the potential to be powerful tools in analysing empirical contact data used in epidemiological studies, but remain untested for use in networks of spatio-temporal associations. 5. As a result, we argue that developments in the statistical analysis of empirical contact data are critical given the ready availability of dynamic network data from bio-logging studies. Furthermore, we encourage improved integration of statistical network approaches into epidemiological research to facilitate the generation of novel modelling frameworks and help extend our understanding of disease transmission in natural populations.

show abstract

Section: Choosing a Modelmentioning

confidence: 99%

The application of statistical network models in disease research

et al. 2017

View full text Add to dashboard Cite

show abstract

“…Morris et al (2008) simply state that “the larger question of how to go about choosing terms wisely is beyond its scope.” Hunter et al (2012) provide a recent review of the ERGM literature (along with several other network models) and discuss model degeneracy, but do not touch on selection of ERGM terms. Model degeneracy refers to the situation in which only a few networks (often the full or empty networks) have appreciable probability given the model and is a challenge in fitting ERGM models.…”

Section: Finite Mixtures Of Ergmsmentioning

confidence: 99%

Role Analysis in Networks Using Mixtures of Exponential Random Graph Models

Salter‐Townshend

Murphy

2015

Journal of Computational and Graphical Statistics

View full text Add to dashboard Cite

A novel and flexible framework for investigating the roles of actors within a network is introduced. Particular interest is in roles as defined by local network connectivity patterns, identified using the ego-networks extracted from the network. A mixture of Exponential-family Random Graph Models is developed for these ego-networks in order to cluster the nodes into roles. We refer to this model as the ego-ERGM. An Expectation-Maximization algorithm is developed to infer the unobserved cluster assignments and to estimate the mixture model parameters using a maximum pseudo-likelihood approximation. The flexibility and utility of the method are demonstrated on examples of simulated and real networks.

show abstract

“…For the literature addressing the problem of selecting K, besides the block-wise edge splitting method of Chen and Lei (2014), a common practice is to use BIC-type criteria (Airoldi et al, 2008;Daudin et al, 2008) or a variational Bayes approach (Latouche et al, 2012;Hunter et al, 2012). An inherently related problem is that of selecting the number of components in mixture models, where the birth-and-death point process of Stephens (2000) and the allocation sampler of Nobile and Fearnside (2007) provide two fully Bayesian approaches in the case where K is finite but unknown.…”

Section: Introductionmentioning

confidence: 99%

How Many Communities Are There?

Saldana

Feng

2017

Journal of Computational and Graphical Statistics

View full text Add to dashboard Cite

Stochastic blockmodels and variants thereof are among the most widely used approaches to community detection for social networks and relational data. A stochastic blockmodel partitions the nodes of a network into disjoint sets, called communities. The approach is inherently related to clustering with mixture models; and raises a similar model selection problem for the number of communities. The Bayesian information criterion (BIC) is a popular solution, however, for stochastic blockmodels, the conditional independence assumption given the communities of the endpoints among different edges is usually violated in practice. In this regard, we propose composite likelihood BIC (CL-BIC) to select the number of communities, and we show it is robust against possible misspecifications in the underlying stochastic blockmodel assumptions. We derive the requisite methodology and illustrate the approach using both simulated and real data. Supplementary materials containing the relevant computer code are available online.

show abstract

Computational Statistical Methods for Social Network Models

Cited by 85 publications

References 92 publications

The application of statistical network models in disease research

The application of statistical network models in disease research

Role Analysis in Networks Using Mixtures of Exponential Random Graph Models

How Many Communities Are There?

Contact Info

Product

Resources

About