A new method to reduce overestimation of thresholds with observational network data

George Berry and Christopher John Cameron

No ctjd6, SocArXiv from Center for Open Science

Abstract: Networks of interdependent nodes support phenomena such as epidemics, product adoption, cascading failure, ecosystem collapse, congestion, and bandwagon effects. We consider the problem of using observational data to estimate the sensitivity of individual nodes to the activation of their network neighbors. We prove that—in the case of binary activation decisions—activation thresholds are impossible to correctly measure for some nodes in virtually all contagion processes on complex networks. This result holds even when each step of the process is observed. Measurement error always produces an overestimate of a node's true activation threshold. We develop a condition for determining which node thresholds are correctly measured and demonstrate that modeling activation thresholds as a function of node-level factors reduces the error compared to existing approaches.

Date: 2017-02-22
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://osf.io/download/58adf4fc594d9001fc145414/

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:osf:socarx:ctjd6

DOI: 10.31219/osf.io/ctjd6

Access Statistics for this paper

More papers in SocArXiv from Center for Open Science
Bibliographic data for series maintained by OSF ().