Modelling Dyadic Interaction with Hawkes Processes

Peter F. Halpin, Paul De Boeck

Research output: Contribution to journalArticle

Abstract

We apply the Hawkes process to the analysis of dyadic interaction. The Hawkes process is applicable to excitatory interactions, wherein the actions of each individual increase the probability of further actions in the near future. We consider the representation of the Hawkes process both as a conditional intensity function and as a cluster Poisson process. The former treats the probability of an action in continuous time via non-stationary distributions with arbitrarily long historical dependency, while the latter is conducive to maximum likelihood estimation using the EM algorithm. We first outline the interpretation of the Hawkes process in the dyadic context, and then illustrate its application with an example concerning email transactions in the work place.

Original languageEnglish (US)
Pages (from-to)793-814
Number of pages22
JournalPsychometrika
Volume78
Issue number4
DOIs
StatePublished - Oct 2013

Fingerprint

Maximum likelihood estimation
Electronic mail
Interaction
Modeling
Workplace
Intensity Function
Electronic Mail
EM Algorithm
Poisson process
Maximum Likelihood Estimation
Transactions
Continuous Time
Context
Interpretation

Keywords

  • dyadic interaction
  • EM algorithm
  • event sampling
  • Hawkes processes
  • maximum likelihood

ASJC Scopus subject areas

  • Psychology(all)
  • Applied Mathematics
  • Medicine(all)

Cite this

Modelling Dyadic Interaction with Hawkes Processes. / Halpin, Peter F.; De Boeck, Paul.

In: Psychometrika, Vol. 78, No. 4, 10.2013, p. 793-814.

Research output: Contribution to journalArticle

Halpin, Peter F. ; De Boeck, Paul. / Modelling Dyadic Interaction with Hawkes Processes. In: Psychometrika. 2013 ; Vol. 78, No. 4. pp. 793-814.
@article{090c1f957d9b44c79330fd7717d53b70,
title = "Modelling Dyadic Interaction with Hawkes Processes",
abstract = "We apply the Hawkes process to the analysis of dyadic interaction. The Hawkes process is applicable to excitatory interactions, wherein the actions of each individual increase the probability of further actions in the near future. We consider the representation of the Hawkes process both as a conditional intensity function and as a cluster Poisson process. The former treats the probability of an action in continuous time via non-stationary distributions with arbitrarily long historical dependency, while the latter is conducive to maximum likelihood estimation using the EM algorithm. We first outline the interpretation of the Hawkes process in the dyadic context, and then illustrate its application with an example concerning email transactions in the work place.",
keywords = "dyadic interaction, EM algorithm, event sampling, Hawkes processes, maximum likelihood",
author = "Halpin, {Peter F.} and {De Boeck}, Paul",
year = "2013",
month = "10",
doi = "10.1007/s11336-013-9329-1",
language = "English (US)",
volume = "78",
pages = "793--814",
journal = "Psychometrika",
issn = "0033-3123",
publisher = "Springer New York",
number = "4",

}

TY - JOUR

T1 - Modelling Dyadic Interaction with Hawkes Processes

AU - Halpin, Peter F.

AU - De Boeck, Paul

PY - 2013/10

Y1 - 2013/10

N2 - We apply the Hawkes process to the analysis of dyadic interaction. The Hawkes process is applicable to excitatory interactions, wherein the actions of each individual increase the probability of further actions in the near future. We consider the representation of the Hawkes process both as a conditional intensity function and as a cluster Poisson process. The former treats the probability of an action in continuous time via non-stationary distributions with arbitrarily long historical dependency, while the latter is conducive to maximum likelihood estimation using the EM algorithm. We first outline the interpretation of the Hawkes process in the dyadic context, and then illustrate its application with an example concerning email transactions in the work place.

AB - We apply the Hawkes process to the analysis of dyadic interaction. The Hawkes process is applicable to excitatory interactions, wherein the actions of each individual increase the probability of further actions in the near future. We consider the representation of the Hawkes process both as a conditional intensity function and as a cluster Poisson process. The former treats the probability of an action in continuous time via non-stationary distributions with arbitrarily long historical dependency, while the latter is conducive to maximum likelihood estimation using the EM algorithm. We first outline the interpretation of the Hawkes process in the dyadic context, and then illustrate its application with an example concerning email transactions in the work place.

KW - dyadic interaction

KW - EM algorithm

KW - event sampling

KW - Hawkes processes

KW - maximum likelihood

UR - http://www.scopus.com/inward/record.url?scp=84884987804&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84884987804&partnerID=8YFLogxK

U2 - 10.1007/s11336-013-9329-1

DO - 10.1007/s11336-013-9329-1

M3 - Article

VL - 78

SP - 793

EP - 814

JO - Psychometrika

JF - Psychometrika

SN - 0033-3123

IS - 4

ER -