Inferential Approaches for Network Analysis: AMEN for Latent Factor Models

2018 ◽  
Vol 27 (2) ◽  
pp. 208-222 ◽  
Author(s):  
Shahryar Minhas ◽  
Peter D. Hoff ◽  
Michael D. Ward
Keyword(s):  
Network Analysis ◽  
Graph Model ◽  
Linear Modeling ◽  
Model Framework ◽  
Modeling Framework ◽  
Random Graph Model ◽  
Exponential Random Graph ◽  
Wide Range ◽  
Out Of Sample ◽  
Latent Factor Models

We introduce a Bayesian approach to conduct inferential analyses on dyadic data while accounting for interdependencies between observations through a set of additive and multiplicative effects (AME). The AME model is built on a generalized linear modeling framework and is thus flexible enough to be applied to a variety of contexts. We contrast the AME model to two prominent approaches in the literature: the latent space model (LSM) and the exponential random graph model (ERGM). Relative to these approaches, we show that the AME approach is (a) to be easy to implement; (b) interpretable in a general linear model framework; (c) computationally straightforward; (d) not prone to degeneracy; (e) captures first-, second-, and third-order network dependencies; and (f) notably outperforms ERGMs and LSMs on a variety of metrics and in an out-of-sample context. In summary, AME offers a straightforward way to undertake nuanced, principled inferential network analysis for a wide range of social science questions.

scholarly journals A theoretical and empirical comparison of the temporal exponential random graph model and the stochastic actor-oriented model

2019 ◽  
Vol 7 (1) ◽  
pp. 20-51 ◽  
Author(s):  
Philip Leifeld ◽  
Skyler J. Cranmer
Keyword(s):  
Network Analysis ◽  
Random Graph ◽  
Graph Model ◽  
Real Data ◽  
The Other ◽  
Exponential Random Graph Model ◽  
Empirical Comparison ◽  
Random Graph Model ◽  
Exponential Random Graph ◽  
Longitudinal Network Analysis

AbstractThe temporal exponential random graph model (TERGM) and the stochastic actor-oriented model (SAOM, e.g., SIENA) are popular models for longitudinal network analysis. We compare these models theoretically, via simulation, and through a real-data example in order to assess their relative strengths and weaknesses. Though we do not aim to make a general claim about either being superior to the other across all specifications, we highlight several theoretical differences the analyst might consider and find that with some specifications, the two models behave very similarly, while each model out-predicts the other one the more the specific assumptions of the respective model are met.

scholarly journals A Case Study of Pyramid Scheme Finance Flow Network Based on Social Network Analysis

2019 ◽  
Vol 11 (16) ◽  
pp. 4370
Author(s):  
Feng ◽  
Sun ◽  
Gong
Keyword(s):  
Social Network ◽  
Social Network Analysis ◽  
Network Analysis ◽  
Large Scale ◽  
Graph Model ◽  
Flow Network ◽  
Random Graph Model ◽  
Exponential Random Graph ◽  
Pyramid Structure ◽  
Pyramid Scheme

(1) Background: The pyramid scheme has caused a large-scale plunder of finances due to the unsustainability of its operating model, which seriously jeopardizes economic development and seriously affects social stability. In various types of networks, the finance flow network plays an extremely important role in the pyramid scheme organization. Through the study of the finance network, the operational nature of pyramid scheme organizations can be effectively explored, and the understanding of pyramid scheme organizations can be deepened to provide a basis for dealing with them. (2) Methods: This paper uses the motifs analysis and exponential random graph model in social network analysis to study the micro-structure and the network construction process of the “5.03” pyramid scheme finance flow network in Hunan, China. (3) Results: The finance flow network is sparse, the microstructure shows a typical pyramid structure; finance flows within the community and eventually flows to the most critical personnel, there is no finance relationship between different communities, and there are few finance relationships between pyramid salesmen of the same level. The inductees are in a key position in the network, which may explain why they are transferred to prosecution.

scholarly journals Inferential Network Analysis with Exponential Random Graph Models

2011 ◽  
Vol 19 (1) ◽  
pp. 66-86 ◽  
Author(s):  
Skyler J. Cranmer ◽  
Bruce A. Desmarais
Keyword(s):  
Network Analysis ◽  
Random Graph ◽  
Graph Model ◽  
Network Data ◽  
Exponential Random Graph Models ◽  
Graph Models ◽  
Random Graph Model ◽  
General Acceptance ◽  
Random Graph Models ◽  
Exponential Random Graph

Methods for descriptive network analysis have reached statistical maturity and general acceptance across the social sciences in recent years. However, methods for statistical inference with network data remain fledgling by comparison. We introduce and evaluate a general model for inference with network data, the Exponential Random Graph Model (ERGM) and several of its recent extensions. The ERGM simultaneously allows both inference on covariates and for arbitrarily complex network structures to be modeled. Our contributions are three-fold: beyond introducing the ERGM and discussing its limitations, we discuss extensions to the model that allow for the analysis of non-binary and longitudinally observed networks and show through applications that network-based inference can improve our understanding of political phenomena.

scholarly journals Modeling customer preferences using multidimensional network analysis in engineering design

2016 ◽  
Vol 2 ◽  
Author(s):  
Mingxian Wang ◽  
Wei Chen ◽  
Yun Huang ◽  
Noshir S. Contractor ◽  
Yan Fu
Keyword(s):  
Network Analysis ◽  
Social Influence ◽  
Social Interactions ◽  
Descriptive Analysis ◽  
Graph Model ◽  
Design Decisions ◽  
Random Graph Model ◽  
Customer Preferences ◽  
Exponential Random Graph ◽  
Product Similarities

Motivated by overcoming the existing utility-based choice modeling approaches, we present a novel conceptual framework of multidimensional network analysis (MNA) for modeling customer preferences in supporting design decisions. In the proposed multidimensional customer–product network (MCPN), customer–product interactions are viewed as a socio-technical system where separate entities of ‘customers’ and ‘products’ are simultaneously modeled as two layers of a network, and multiple types of relations, such as consideration and purchase, product associations, and customer social interactions, are considered. We first introduce a unidimensional network where aggregated customer preferences and product similarities are analyzed to inform designers about the implied product competitions and market segments. We then extend the network to a multidimensional structure where customer social interactions are introduced for evaluating social influence on heterogeneous product preferences. Beyond the traditional descriptive analysis used in network analysis, we employ the exponential random graph model (ERGM) as a unified statistical inference framework to interpret complex preference decisions. Our approach broadens the traditional utility-based logit models by considering dependency among complex customer–product relations, including the similarity of associated products, ‘irrationality’ of customers induced by social influence, nested multichoice decisions, and correlated attributes of customers and products.

scholarly journals Symptom and Age Homophilies in SARS-CoV-2 Transmission Networks during the Early Phase of the Pandemic in Japan

Biology ◽  
2021 ◽  
Vol 10 (6) ◽  
pp. 499
Author(s):  
Ali Andalibi ◽  
Naoru Koizumi ◽  
Meng-Hao Li ◽  
Abu Bakkar Siddique
Keyword(s):  
Early Stage ◽  
Graph Model ◽  
Initial Response ◽  
Contact Tracing ◽  
Pcr Analysis ◽  
Exponential Random Graph Model ◽  
Transmission Networks ◽  
Random Graph Model ◽  
Healthcare Settings ◽  
Exponential Random Graph

Kanagawa and Hokkaido were affected by COVID-19 in the early stage of the pandemic. Japan’s initial response included contact tracing and PCR analysis on anyone who was suspected of having been exposed to SARS-CoV-2. In this retrospective study, we analyzed publicly available COVID-19 registry data from Kanagawa and Hokkaido (n = 4392). Exponential random graph model (ERGM) network analysis was performed to examine demographic and symptomological homophilies. Age, symptomatic, and asymptomatic status homophilies were seen in both prefectures. Symptom homophilies suggest that nuanced genetic differences in the virus may affect its epithelial cell type range and can result in the diversity of symptoms seen in individuals infected by SARS-CoV-2. Environmental variables such as temperature and humidity may also play a role in the overall pathogenesis of the virus. A higher level of asymptomatic transmission was observed in Kanagawa. Moreover, patients who contracted the virus through secondary or tertiary contacts were shown to be asymptomatic more frequently than those who contracted it from primary cases. Additionally, most of the transmissions stopped at the primary and secondary levels. As expected, significant viral transmission was seen in healthcare settings.

scholarly journals A Case-Control Study to Identify Community Venues Associated with Genetically-clustered, Multidrug-resistant Tuberculosis Disease in Lima, Peru

2018 ◽  
Vol 68 (9) ◽  
pp. 1547-1555
Author(s):  
David P Bui ◽  
Eyal Oren ◽  
Denise J Roe ◽  
Heidi E Brown ◽  
Robin B Harris ◽  
...  
Keyword(s):  
Graph Model ◽  
Multidrug Resistant ◽  
Random Graph Model ◽  
Tuberculosis Transmission ◽  
Resistant Tuberculosis ◽  
Multidrug Resistant Tuberculosis ◽  
Exponential Random Graph ◽  
The Social ◽  
Community Segregation ◽  
Tuberculosis Disease

Abstract Background The majority of tuberculosis transmission occurs in community settings. Our primary aim in this study was to assess the association between exposure to community venues and multidrug-resistant (MDR) tuberculosis. Our secondary aim was to describe the social networks of MDR tuberculosis cases and controls. Methods We recruited laboratory-confirmed MDR tuberculosis cases and community controls that were matched on age and sex. Whole-genome sequencing was used to identify genetically clustered cases. Venue tracing interviews (nonblinded) were conducted to enumerate community venues frequented by participants. Logistic regression was used to assess the association between MDR tuberculosis and person-time spent in community venues. A location-based social network was constructed, with respondents connected if they reported frequenting the same venue, and an exponential random graph model (ERGM) was fitted to model the network. Results We enrolled 59 cases and 65 controls. Participants reported 729 unique venues. The mean number of venues reported was similar in both groups (P = .92). Person-time in healthcare venues (adjusted odds ratio [aOR] = 1.67, P = .01), schools (aOR = 1.53, P < .01), and transportation venues (aOR = 1.25, P = .03) was associated with MDR tuberculosis. Healthcare venues, markets, cinemas, and transportation venues were commonly shared among clustered cases. The ERGM indicated significant community segregation between cases and controls. Case networks were more densely connected. Conclusions Exposure to healthcare venues, schools, and transportation venues was associated with MDR tuberculosis. Intervention across the segregated network of case venues may be necessary to effectively stem transmission.

Sign in / Sign up

Export Citation Format

Share Document