Modeling Networks When Data Is Missing or Sampled

• Activity Number: 218
• Type: Invited
• Date/Time: Monday, August 5, 2013 : 2:00 PM to 3:50 PM
• Sponsor: IMS
• Author(s): Mark Stephen Handcock*+ and Krista J. Gile and Ian Fellows
• Companies: University of California - Los Angeles and University of Massachusetts, Amherst and University of California - Los Angeles
• Keywords: networks ; network sampling ; likelihood ; exponential families ; missing data ; survey sampling

Abstract

Network models are widely used to represent relational information among interacting units and the structural implications of these relations.

Most inference for social network models assumes that the presence or absence of all possible links is observed, that the information is completely reliable, and that there are no measurement errors. This is clearly not true in practice, as much network data is collected though sample surveys. In addition even if a census of a population is attempted, individuals and links between individuals are missed.

We develop the conceptual and computational theory for inference based on sampled network information. We consider inference within the likelihood framework based on both conventional and adaptive network designs.

We motivate and illustrate these ideas by analyzing the effect of link-tracing sampling designs on a collaboration network, and of missing data in a friendship network among adolescents.