Identifying the Influential Variable using Centrality Measure: A Case of Multivariate Time Series

• Main Authors: Nur Syahidah, Yusoff, Shamshuritawati, Sharif
• Format: Conference or Workshop Item
• Language: English; English
• Published: American Scientific Publishers 2016
• Subjects: Q Science (General)
• Online Access: http://umpir.ump.edu.my/id/eprint/14012/1/Identifying%20the%20Influential%20Variable%20using%20Centrality.pdf; http://umpir.ump.edu.my/id/eprint/14012/7/fist-2016-syahidah-Identifying%20the%20Influential%20Centrality.pdf; http://umpir.ump.edu.my/id/eprint/14012/

Analysing the structure of multivariate system has been an important part in reliability analysis especially in identifying the influential variables. The complexity of the analysis increases when high dimensional data involved. To simplify the information in multivariate system, a network topology which is based on an Escoufier’s RV-coefficient is constructed and centrality measure will be used to interpret the network. Statistically, RV-coefficient is a multivariate generalization of the squared Pearson correlation coefficient. An example in finance industry will be discussed to illustrate the structure of network topology and a recommendation will be presented.