Google Maps E-mail RSS
formats

NCASD lab members to present at 2014 Sunbelt Conference

Published on February 19, 2014 by

NCASD lab members will be presenting at the 2014 Sunbelt Conference in St. Pete Beach Florida from February 18-23.

Sean Fitzhugh is presenting his joint work with Carter Butts on linking shared social contexts to shared structural contexts.  They introduce a family of techniques that relate subgroup co-membership to shared individual features in order to determine how shared attributes may drive group formation.

Ben Gibson is presenting his work with Yue Yu, Zack Almquist, and Carter Butts on a scalable approach to approximate TERGM inference for certain dynamic network regression families, that can scale to arbitrarily large populations.

Emma Smith is presenting joint work with Chris Marcum, Adam Boessen, Zack Almquist, John Hipp, Nicholas Nagle, and Carter Butts on the relationship of age to personal network size, relational multiplexity, and proximity to alters among rural and urban populations in the western United States.

Xuhong Zhang is presenting her work with Carter Butts on a new method for inferring relationships by exploiting the distributional and spectral structure of activity correlation within dyads. They demonstrate methodology via an application to detection of friendship and group co-membership using data from mobile devices.

Yue Yu is presenting her joint work with Emma Smith and Carter Butts on the evaluation of different retrospective life history designs through an examination of the accumulation of missingness as a function of time prior to interview, and the investigation of the impact of this missing data on model-based imputation of the state of the network at prior time points via conditional ERGM prediction.

Lab PI Butts will be presenting work on novel methods for fitting ERGMs to multiple networks – e.g., sets of networks arising from population or experimental studies, or network time series.  These new methods are scalable to collections of hundreds or even thousands of networks, with minimal increase in computational cost.

Good luck to all!

For more information, see here.

 
 Share on Facebook Share on Twitter Share on Reddit Share on LinkedIn
No Comments  comments 
© Copyright NCASD 2012