Open Access
June 2018 Loglinear model selection and human mobility
Adrian Dobra, Reza Mohammadi
Ann. Appl. Stat. 12(2): 815-845 (June 2018). DOI: 10.1214/18-AOAS1164


Methods for selecting loglinear models were among Steve Fienberg’s research interests since the start of his long and fruitful career. After we dwell upon the string of papers focusing on loglinear models that can be partly attributed to Steve’s contributions and influential ideas, we develop a new algorithm for selecting graphical loglinear models that is suitable for analyzing hyper-sparse contingency tables. We show how multi-way contingency tables can be used to represent patterns of human mobility. We analyze a dataset of geolocated tweets from South Africa that comprises $46$ million latitude/longitude locations of $476\mbox{,}601$ Twitter users that is summarized as a contingency table with $214$ variables.


Download Citation

Adrian Dobra. Reza Mohammadi. "Loglinear model selection and human mobility." Ann. Appl. Stat. 12 (2) 815 - 845, June 2018.


Received: 1 November 2017; Revised: 1 March 2018; Published: June 2018
First available in Project Euclid: 28 July 2018

zbMATH: 06980477
MathSciNet: MR3834287
Digital Object Identifier: 10.1214/18-AOAS1164

Keywords: Bayesian structural learning , Birth–death processes , Contingency tables , graphical models , human mobility , Model selection , pseudo-likelihood

Rights: Copyright © 2018 Institute of Mathematical Statistics

Vol.12 • No. 2 • June 2018
Back to Top