Bayesian exponential random graph modelling of interhospital patient referral networks

Caimo, Alberto; Pallotti, Francesca; Lomi, Alessandro

doi:10.1002/sim.7301

Cited by 25 publications

(16 citation statements)

References 71 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Exponential Random Graph Models (ERGMs) are statistical models of network structure, permitting inferences about how network ties are patterned [52]. ERGMs have been applied to several fields such as economics [53], sociology [54], political sciences [55], international relations [56], medicine [57] and public health [58] with varied application ranging from modeling micro-blog networks [59], studying relational coordination among healthcare organizations [60] to strategic management research [61]. Social network models too have attracted considerable attention from physicists [62,63] and have been pivotal in the development of interdisciplinary perspectives [64].…”

Section: Exponential Random Graphsmentioning

confidence: 99%

Patent citation network analysis: A perspective from descriptive statistics and ERGMs

2020

View full text Add to dashboard Cite

Patent Citation Analysis has been gaining considerable traction over the past few decades. In this paper, we collect extensive information on patents and citations and provide a perspective of citation network analysis of patents from a statistical viewpoint. We identify and analyze the most cited patents, the most innovative and the highly cited companies along with the structural properties of the network by providing in-depth descriptive analysis. Furthermore, we employ Exponential Random Graph Models (ERGMs) to analyze the citation networks. ERGMs enables understanding the social perspectives of a patent citation network which has not been studied earlier. We demonstrate that social properties such as homophily (the inclination to cite patents from the same country or in the same language) and transitivity (the inclination to cite references’ references) together with the technicalities of the patents (e.g., language, categories), has a significant effect on citations. We also provide an in-depth analysis of citations for sectors in patents and how it is affected by the size of the same. Overall, our paper delves into European patents with the aim of providing new insights and serves as an account for fitting ERGMs on large networks and analyzing them. ERGMs help us model network mechanisms directly, instead of acting as a proxy for unspecified dependence and relationships among the observations.

show abstract

Section: Exponential Random Graphsmentioning

confidence: 99%

Patent citation network analysis: A perspective from descriptive statistics and ERGMs

2020

View full text Add to dashboard Cite

show abstract

“…In recent years, ERGMs have found applications in empirical research in a wide range of scientific fields. Recent examples include the study of large friendship networks (Goodreau, 2007), genetic and metabolic networks (Saul and Filkov, 2007), disease transmission networks (Groendyke et al, 2012), conflict networks in the international system (Cranmer and Desmarais, 2011), the structure of ancient networks in various of archaeological settings (Amati et al, 2019), the structural comparison of protein structure networks (Grazioli et al, 2019), the effects of functional integration and functional segregation in brain functional connectivity networks (Simpson et al, 2011;Sinke et al, 2016;Obando and De Vico Fallani, 2017), and the impact of endogenous network effects on the formation of interhospital patient referral networks (Caimo et al, 2017). While addressing very different problems in different empirical settings, what these studies have in common is a clear methodological commitment to modeling network mechanisms directly via parametric effects, rather than just attempting to "control for" unspecified dependence among the observations (e.g., via latent structure).…”

Section: Exponential-family Random Graph Models (Ergms)mentioning

confidence: 99%

Finite Mixtures of ERGMs for Modeling Ensembles of Networks

2022

View full text Add to dashboard Cite

Ensembles of networks arise in many scientific fields, but there are relatively few statistical tools for inferring their generative processes, particularly in the presence of both dyadic dependence and cross-graph heterogeneity. To address this gap, we propose characterizing network ensembles via finite mixtures of exponential family random graph models (ERGMs), a class of parametric statistical models that has been successful in explicitly modeling the complex stochastic processes that govern the structure of edges in a network. Our proposed modeling framework can also be used for applications such as model-based clustering of ensembles of networks and density estimation for complex graph distributions. We develop a joint approach to estimate the number of mixture components and identify cluster-specific parameters simultaneously as well as to obtain an identified model under the Bayesian paradigm. Specifically, we develop a Metropoliswithin-Gibbs algorithm to perform Bayesian inference, and estimate the number of mixture components using a strategy of deliberate overfitting with sparse priors that removes excess components during MCMC. As the true ERGM likelihood is generally intractable for model specifications with dyadic dependence terms, we consider two tractable approximations (pseudolikelihood and adjusted pseudolikelihood) to facilitate efficient statistical inference. We run simulation studies to compare the performance of these two approximations with respect to multiple metrics, showing conditions under which both are useful. We demonstrate the utility of the proposed approach using an ensemble of political co-voting networks among U.S. Senators and an ensemble of brain functional connectivity networks.

show abstract

“…We use ERGMs in their Bayesian form. Bayesian ERGMS (BERGMs) offer the inherent features of Bayesian approaches, such as the intuitive interpretation of parameter estimates as posterior distributions, together with considerable promise in alleviating common ERGM problems such as computational tractability, degeneracy and interpreting parameter estimates (Caimo, Pallotti, and Lomi 2017). To ensure the model adequately represents endogenous processes in our data, we perform goodness-of-fit tests (see supplementary materials).…”

Section: Identifying Divisions In Swiss Water Governancementioning

confidence: 99%

Information exchange in governance networks—Who brokers across political divides?

Angst

Brandenberger

2021

Governance

View full text Add to dashboard Cite

The exchange of information among organizations is the lifeblood of governance networks. It is a necessary condition for successful governance. Political divides between organizations often impede information exchange. We ask which organizations are most likely to broker information across political divides. We consider survey (n = 312) evidence of technical information exchange in Swiss water governance. Bayesian exponential random graph modeling results show that scientific organizations play crucial roles for cross-divide brokerage. To a lesser extent, this also holds for higher-level governmental agencies. Participation in policy forums is associated with a higher likelihood of brokering across political divides. There is however no clear benefit to participating in more than two or three forums. We conclude that an active role played by scientific organizations is the most promising avenue to increase information flow across political divides. Beyond this, we suggest setting up carefully constructed exchange forums and selectively engaging moderate members of subgroups.

show abstract

Bayesian exponential random graph modelling of interhospital patient referral networks

Cited by 25 publications

References 71 publications

Patent citation network analysis: A perspective from descriptive statistics and ERGMs

Patent citation network analysis: A perspective from descriptive statistics and ERGMs

Finite Mixtures of ERGMs for Modeling Ensembles of Networks

Information exchange in governance networks—Who brokers across political divides?

Contact Info

Product

Resources

About