Details
| Original language | English |
|---|---|
| Pages (from-to) | 1337-1364 |
| Number of pages | 28 |
| Journal | Statistical Methods & Applications |
| Volume | 30 |
| Issue number | 5 |
| Early online date | 15 Sept 2021 |
| Publication status | Published - Dec 2021 |
Abstract
An important problem in network analysis is the online detection of anomalous behaviour. In this paper, we introduce a network surveillance method bringing together network modelling and statistical process control. Our approach is to apply multivariate control charts based on exponential smoothing and cumulative sums in order to monitor networks generated by temporal exponential random graph models (TERGM). The latter allows us to account for temporal dependence while simultaneously reducing the number of parameters to be monitored. The performance of the considered charts is evaluated by calculating the average run length and the conditional expected delay for both simulated and real data. To justify the decision of using the TERGM to describe network data, some measures of goodness of fit are inspected. We demonstrate the effectiveness of the proposed approach by an empirical application, monitoring daily flights in the United States to detect anomalous patterns.
Keywords
- MCUSUM, MEWMA, Multivariate Control Charts, Network Modelling, Network Monitoring, Statistical Process Control, TERGM
ASJC Scopus subject areas
- Mathematics(all)
- Statistics and Probability
- Decision Sciences(all)
- Statistics, Probability and Uncertainty
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: Statistical Methods & Applications, Vol. 30, No. 5, 12.2021, p. 1337-1364.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - Online network monitoring
AU - Malinovskaya, Anna
AU - Otto, Philipp
N1 - Funding Information: We thank participants of several research seminars at the Leibniz University Hannover for their inspiring remarks and particularly Thomas Cope for his valuable comments and suggestions. We thank the Guest Editor and the two Reviewers for their comprehensive suggestions and comments. The results presented here were carried out on the cluster system at the Leibniz University of Hannover. The project is funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) - 412992257.
PY - 2021/12
Y1 - 2021/12
N2 - An important problem in network analysis is the online detection of anomalous behaviour. In this paper, we introduce a network surveillance method bringing together network modelling and statistical process control. Our approach is to apply multivariate control charts based on exponential smoothing and cumulative sums in order to monitor networks generated by temporal exponential random graph models (TERGM). The latter allows us to account for temporal dependence while simultaneously reducing the number of parameters to be monitored. The performance of the considered charts is evaluated by calculating the average run length and the conditional expected delay for both simulated and real data. To justify the decision of using the TERGM to describe network data, some measures of goodness of fit are inspected. We demonstrate the effectiveness of the proposed approach by an empirical application, monitoring daily flights in the United States to detect anomalous patterns.
AB - An important problem in network analysis is the online detection of anomalous behaviour. In this paper, we introduce a network surveillance method bringing together network modelling and statistical process control. Our approach is to apply multivariate control charts based on exponential smoothing and cumulative sums in order to monitor networks generated by temporal exponential random graph models (TERGM). The latter allows us to account for temporal dependence while simultaneously reducing the number of parameters to be monitored. The performance of the considered charts is evaluated by calculating the average run length and the conditional expected delay for both simulated and real data. To justify the decision of using the TERGM to describe network data, some measures of goodness of fit are inspected. We demonstrate the effectiveness of the proposed approach by an empirical application, monitoring daily flights in the United States to detect anomalous patterns.
KW - MCUSUM
KW - MEWMA
KW - Multivariate Control Charts
KW - Network Modelling
KW - Network Monitoring
KW - Statistical Process Control
KW - TERGM
UR - http://www.scopus.com/inward/record.url?scp=85114943252&partnerID=8YFLogxK
U2 - 10.1007/s10260-021-00589-z
DO - 10.1007/s10260-021-00589-z
M3 - Article
VL - 30
SP - 1337
EP - 1364
JO - Statistical Methods & Applications
JF - Statistical Methods & Applications
SN - 1613-981X
IS - 5
ER -