TY - GEN
T1 - Detection of rare events in multidimensional financial datasets with zonoid depth functions
AU - Golbayani, Parisa
AU - Bozdog, Dragos
N1 - Publisher Copyright:
© 2017 IEEE.
PY - 2017/7/1
Y1 - 2017/7/1
N2 - The analysis of occurrence of rare events along with their significant negative economic impacts has been a challenging topic in recent years. In this study, we construct a framework around the development of new techniques that extracts information out of big datasets that spans the areas of data science and financial model analysis. In particular, we investigate methods to reduce the complexity of the analysis, map large amounts of data and construct a model to detect rare events on multivariate financial datasets. Zonoid depth functions are implemented and classification criteria based on convex hulls are constructed. The results identify rare events in high-frequency transaction level data and the distribution of these rare events for the period investigated indicates a significant higher concentration of rare events in the first trading hours of each trading day.
AB - The analysis of occurrence of rare events along with their significant negative economic impacts has been a challenging topic in recent years. In this study, we construct a framework around the development of new techniques that extracts information out of big datasets that spans the areas of data science and financial model analysis. In particular, we investigate methods to reduce the complexity of the analysis, map large amounts of data and construct a model to detect rare events on multivariate financial datasets. Zonoid depth functions are implemented and classification criteria based on convex hulls are constructed. The results identify rare events in high-frequency transaction level data and the distribution of these rare events for the period investigated indicates a significant higher concentration of rare events in the first trading hours of each trading day.
UR - http://www.scopus.com/inward/record.url?scp=85046160581&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85046160581&partnerID=8YFLogxK
U2 - 10.1109/SSCI.2017.8280978
DO - 10.1109/SSCI.2017.8280978
M3 - Conference contribution
AN - SCOPUS:85046160581
T3 - 2017 IEEE Symposium Series on Computational Intelligence, SSCI 2017 - Proceedings
SP - 1
EP - 6
BT - 2017 IEEE Symposium Series on Computational Intelligence, SSCI 2017 - Proceedings
T2 - 2017 IEEE Symposium Series on Computational Intelligence, SSCI 2017
Y2 - 27 November 2017 through 1 December 2017
ER -