2015
DOI: 10.1016/j.ejor.2014.12.041
|View full text |Cite
|
Sign up to set email alerts
|

Clustering financial time series: New insights from an extended hidden Markov model

Abstract: In recent years, large amounts of financial data have become available for analysis. We propose exploring returns from 21 European stock markets by model-based clustering of regime switching models. These econometric models identify clusters of time series with similar dynamic patterns and moreover allow relaxing assumptions of existing approaches, such as the assumption of conditional Gaussian returns. The proposed model handles simultaneously the heterogeneity across stock markets and over time, i.e., time-c… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

1
44
0

Year Published

2017
2017
2024
2024

Publication Types

Select...
6
2
1

Relationship

0
9

Authors

Journals

citations
Cited by 91 publications
(45 citation statements)
references
References 58 publications
1
44
0
Order By: Relevance
“…There are also excellent studies that focus on improving the existing HMM based clustering framework, such as using Dynamic Time Warping to bootstrap the process of fitting HMMs [26] and introducing the Bayesian approach to HMM clustering [27]. A relatively similar work to ours is seen in J. G. Dias et al [28] where the authors derive insights about groups of stock markets, and how the regime switching dynamics differ across these groups, using an extended multilevel HMM. Our work further looks into how different groups of customers behave under the same global regime switching characteristics, as well as how they switch between regimes under the same global behavior patterns.…”
Section: Related Workmentioning
confidence: 99%
“…There are also excellent studies that focus on improving the existing HMM based clustering framework, such as using Dynamic Time Warping to bootstrap the process of fitting HMMs [26] and introducing the Bayesian approach to HMM clustering [27]. A relatively similar work to ours is seen in J. G. Dias et al [28] where the authors derive insights about groups of stock markets, and how the regime switching dynamics differ across these groups, using an extended multilevel HMM. Our work further looks into how different groups of customers behave under the same global regime switching characteristics, as well as how they switch between regimes under the same global behavior patterns.…”
Section: Related Workmentioning
confidence: 99%
“…Clustering asset returns into time periods with similar behavior is dierent from other types of clustering, such as k-means, due to the time dependence (Dias et al, 2015). In machine learning, the task of inferring a function to describe a hidden structure from unlabeled data is called unsupervised learning.…”
Section: Data Modelmentioning
confidence: 99%
“…Recently Dias et al (2015) using an extended Hidden Markov Model, show the presence of three regimes: the bull, the bear and a stable regime with the stable regime occurring most frequently. One can use this method to detect regime changes.…”
Section: Concluding Commentsmentioning
confidence: 99%
“…Changes from stable to bull will count as hitting upper threshold and changes from stable to bear will count as hitting lower threshold. The full development in our paper can then be carried out by combining our method with that of Dias et al (2015). It will detect contagion in the sense of entering bull or bear states in one market drives the same behavior in the other.…”
Section: Concluding Commentsmentioning
confidence: 99%