Networks Evolving Step by Step: Statistical Analysis of Dyadic Event Data

Brandes, Ulrik; Lerner, Jürgen; Snijders, Tom A. B.

doi:10.1109/asonam.2009.28

Cited by 113 publications

(117 citation statements)

References 18 publications

Supporting

Mentioning

116

Contrasting

Unclassified

Order By: Relevance

“…This is a competing risks event history model, predicting both the occurrence and the quality of the link (cf. Brandes et al 2009). It can be estimated with a multilevel multinomial regression model.…”

Section: Conclusion and Discussionmentioning

confidence: 99%

“…As a consequence, the estimated overall piecewise hazard function, which concatenates all piece-wise constant hazards, is different from hazard functions in event history models, which do not reset the clock during the process. Butts' method was extended by Brandes et al (2009) to account for competing risks.…”

Section: Event History Models Applied To Social Networkmentioning

confidence: 99%

See 1 more Smart Citation

Networks of action and events over time. A multilevel discrete-time event history model for longitudinal network data

Nooy

2011

Social Networks

View full text Add to dashboard Cite

Longitudinal network data recording the moment at which ties appear, change, or disappear are increasingly available. Event history models can be used to analyze the dynamics of time-stamped network data. This paper adapts the discrete-time event history model to social network data. A discrete-time event history model can easily incorporate a multilevel design and time-varying covariates. A multilevel design is needed to account for dependencies among ties and vertices, which should not be ignored in a small longitudinal network. Time-varying covariates are required to analyze network effects, that is, the impact of previous ties. In addition, a discrete-time event history model handles constraints on who can act or who can be acted upon in a straightforward way. The model can be estimated with multilevel logistic regression analysis, which is illustrated by an application to book reviews, so network evolution can be analyzed with a fairly standard statistical tool.Keywords: network dynamics; discrete-time event history models; multilevel logistic regression analysis; longitudinal social networks; book reviewing.In the analysis of longitudinal social networks, relatively little attention has been paid to the timing of relational events, namely the moments at which ties appear, change, or disappear. Initially, this kind of data was not available because longitudinal data were collected by means of repeatedly administered surveys (Katz and Proctor 1959: 318; Runger and Wasserman 1979: 144). As a consequence, models for network dynamics

show abstract

Section: Conclusion and Discussionmentioning

confidence: 99%

Section: Event History Models Applied To Social Networkmentioning

confidence: 99%

Networks of action and events over time. A multilevel discrete-time event history model for longitudinal network data

Nooy

2011

Social Networks

View full text Add to dashboard Cite

show abstract

“…For a number of reasons a standard econometric model for continuous time series data is preferable as in event dyads research (Brandes, Lerner, & Snijders, 2009). First, our focus is on the extent to which parties take pro or con stances to issues rather than on ordinal values of positions, and on the precise degree of support or criticism among parties, thus on continuous tie value rather than on the discrete presence or absence of a tie.…”

Section: Towards a Two-mode Dynamic Network Adjustment Modelmentioning

confidence: 99%

Adjustment of issue positions based on network strategies in an election campaign: A two-mode network autoregression model with cross-nested random effects

Kleinnijenhuis¹,

Nooy

2013

Social Networks

View full text Add to dashboard Cite

“…In contrast, we focus on social network data that can be viewed as a set of relational events [5,4]. Each event is an instantaneous or finite-duration action involving two or more entities.…”

Section: Introductionmentioning

confidence: 99%

Modeling relational events via latent classes

DuBois

Smyth

2010

Proceedings of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

View full text Add to dashboard Cite

Many social networks can be characterized by a sequence of dyadic interactions between individuals. Techniques for analyzing such events are of increasing interest. In this paper, we describe a generative model for dyadic events, where each event arises from one of C latent classes, and the properties of the event (sender, recipient, and type) are chosen from distributions over these entities conditioned on the chosen class. We present two algorithms for inference in this model: an expectation-maximization algorithm as well as a Markov chain Monte Carlo procedure based on collapsed Gibbs sampling. To analyze the model's predictive accuracy, the algorithms are applied to multiple real-world data sets involving email communication, international political events, and animal behavior data.

show abstract

Networks Evolving Step by Step: Statistical Analysis of Dyadic Event Data

Cited by 113 publications

References 18 publications

Networks of action and events over time. A multilevel discrete-time event history model for longitudinal network data

Networks of action and events over time. A multilevel discrete-time event history model for longitudinal network data

Adjustment of issue positions based on network strategies in an election campaign: A two-mode network autoregression model with cross-nested random effects

Modeling relational events via latent classes

Contact Info

Product

Resources

About