network-modeling

Modeling the Visibility Distribution for Respondent-Driven Sampling with Application to Population Size Estimation

We develop a method for improved estimation of a participant’s inclusion probability based on their network size (degree) as well as other information.

Causal Inference over Stochastic Networks

Claiming causal inferences in network settings necessitates careful consideration of the often complex dependency between outcomes for actors. Of particular importance are treatment spillover or outcome interference effects. We consider causal …

Exposure Notification System activity as a leading indicator for SARS-COV-2 caseload forecasting

We show that smartphone based Exposure Notification systems can significantly improve the accuracy of short-term forecasting of COVID-19 caseloads.

Understanding Networks with Exponential-family Random Network Models

We argue that a new class of exponential-family models for networks is more appriopriate when some of the covariates are stochastic.

Modeling of networked populations when data is sampled or missing

We discuss network modeling with a novel exponential-family class of models when the network has some stochastic covariates and is only partially observed.

A Practical Revealed Preference Model for Separating Preferences and Availability Effects in Marriage Formation

Many demographic problems require models for partnership formation. We consider a model for matchings within a bipartite population where individuals have utility for people based on observed and unobserved characteristics. It represents both the …

Practical Network Modeling via Tapered Exponential-family Random Graph Models

Exponential-family Random Graph Models (ERGMs) have long been at the forefront of the analysis of relational data. The exponential-family form allows complex network dependencies to be represented. Models in this class are interpretable, flexible and …

Comparing the real-world performance of exponential-family random graph models and latent order logistic models for social network analysis

We assess the real-world performance of Latent Order Logistic models (LOLOG) when applied to typical networks modelled by researchers by comparing them to Exponential-family random graph models (ERGMs). We demonstrate that the LOLOG models are, in …

Comment on "On Nonparametric Bayes Modeling of Populations of Networks"

This is an invited discussion of `Nonparametric Bayes Modeling of Populations of Networks` by Daniele Durante, David B. Dunson and Joshua T. Vogelstein.

Removing Phase Transitions from Gibbs Measures

Gibbs measures are a fundamental class of distributions for the analysis of high dimensional data. Phase transitions, which are also known as degeneracy in the network science literature, are an emergent property of these models that well describe …