Publication

Multiple imputation for longitudinal network data

Krause, R. W., Huisman, M. & Snijders, T. A. B., 2018, In : Statistica Applicata - Italian Journal of Applied Statistics. 30, 1, p. 33-57 25 p.

Research output: Contribution to journalArticleAcademicpeer-review

Copy link to clipboard

Documents

  • Multiple imputation for longitudinal network data

    Final publisher's version, 231 KB, PDF document

DOI

Missing data on network ties are a fundamental problem for network analysis.
The biases induced by missing edge data are widely acknowledged. In this paper, we present a new method with two variants to handle missing data due to actor non-response in the framework of stochastic actor-oriented models (SAOMs). The proposed method imputes missing tie variables in the first wave either by using a Bayesian exponential random graph model (BERGMs) or a stationary SAOM and imputes missing tie variables in later waves utilizing a SAOM. The proposed method is compared to the standard SAOM missing data treatment as well as recently proposed methods. The multiple imputation procedure provided more reliable point estimates than the default treatment. The results have relevant implications for the analysis of network dynamics under missing data.
Original languageEnglish
Pages (from-to)33-57
Number of pages25
JournalStatistica Applicata - Italian Journal of Applied Statistics
Volume30
Issue number1
Publication statusPublished - 2018

ID: 64341886