High-speed detection of emergent market clustering via an unsupervised parallel genetic algorithm
We implement a master-slave parallel genetic algorithm with a bespoke log-likelihood fitness function to identify emergent clusters within price evolutions. We use graphics processing units (GPUs) to implement a parallel genetic algorithm and visualise the results using disjoint minimal spanning tre...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Academy of Science of South Africa
2016-02-01
|
| Series: | South African Journal of Science |
| Subjects: | |
| Online Access: | https://www.sajs.co.za/article/view/4149 |
| _version_ | 1819295585517174784 |
|---|---|
| author | Dieter Hendricks Tim Gebbie Diane Wilcox |
| author_facet | Dieter Hendricks Tim Gebbie Diane Wilcox |
| author_sort | Dieter Hendricks |
| collection | DOAJ |
| description | We implement a master-slave parallel genetic algorithm with a bespoke log-likelihood fitness function to identify emergent clusters within price evolutions. We use graphics processing units (GPUs) to implement a parallel genetic algorithm and visualise the results using disjoint minimal spanning trees. We demonstrate that our GPU parallel genetic algorithm, implemented on a commercially available general purpose GPU, is able to recover stock clusters in sub-second speed, based on a subset of stocks in the South African market. This approach represents a pragmatic choice for low-cost, scalable parallel computing and is significantly faster than a prototype serial implementation in an optimised C-based fourth-generation programming language, although the results are not directly comparable because of compiler differences. Combined with fast online intraday correlation matrix estimation from high frequency data for cluster identification, the proposed implementation offers cost-effective, near-real-time risk assessment for financial practitioners. |
| first_indexed | 2024-12-24T04:44:33Z |
| format | Article |
| id | doaj.art-57bac2a09ab24e92bcb6730cc38978d9 |
| institution | Directory Open Access Journal |
| issn | 1996-7489 |
| language | English |
| last_indexed | 2024-12-24T04:44:33Z |
| publishDate | 2016-02-01 |
| publisher | Academy of Science of South Africa |
| record_format | Article |
| series | South African Journal of Science |
| spelling | doaj.art-57bac2a09ab24e92bcb6730cc38978d92022-12-21T17:14:42ZengAcademy of Science of South AfricaSouth African Journal of Science1996-74892016-02-011121/29910.17159/sajs.2016/201403404149High-speed detection of emergent market clustering via an unsupervised parallel genetic algorithmDieter Hendricks0Tim Gebbie1Diane Wilcox2School of Computer Science and Applied Mathematics, University of the Witwatersrand, Johannesburg, South AfricaSchool of Computer Science and Applied Mathematics, University of the Witwatersrand, Johannesburg, South AfricaSchool of Computer Science and Applied Mathematics, University of the Witwatersrand, Johannesburg, South AfricaWe implement a master-slave parallel genetic algorithm with a bespoke log-likelihood fitness function to identify emergent clusters within price evolutions. We use graphics processing units (GPUs) to implement a parallel genetic algorithm and visualise the results using disjoint minimal spanning trees. We demonstrate that our GPU parallel genetic algorithm, implemented on a commercially available general purpose GPU, is able to recover stock clusters in sub-second speed, based on a subset of stocks in the South African market. This approach represents a pragmatic choice for low-cost, scalable parallel computing and is significantly faster than a prototype serial implementation in an optimised C-based fourth-generation programming language, although the results are not directly comparable because of compiler differences. Combined with fast online intraday correlation matrix estimation from high frequency data for cluster identification, the proposed implementation offers cost-effective, near-real-time risk assessment for financial practitioners.https://www.sajs.co.za/article/view/4149unsupervised clusteringgenetic algorithmsparallel algorithmsfinancial data processingmaximum likelihood clustering |
| spellingShingle | Dieter Hendricks Tim Gebbie Diane Wilcox High-speed detection of emergent market clustering via an unsupervised parallel genetic algorithm South African Journal of Science unsupervised clustering genetic algorithms parallel algorithms financial data processing maximum likelihood clustering |
| title | High-speed detection of emergent market clustering via an unsupervised parallel genetic algorithm |
| title_full | High-speed detection of emergent market clustering via an unsupervised parallel genetic algorithm |
| title_fullStr | High-speed detection of emergent market clustering via an unsupervised parallel genetic algorithm |
| title_full_unstemmed | High-speed detection of emergent market clustering via an unsupervised parallel genetic algorithm |
| title_short | High-speed detection of emergent market clustering via an unsupervised parallel genetic algorithm |
| title_sort | high speed detection of emergent market clustering via an unsupervised parallel genetic algorithm |
| topic | unsupervised clustering genetic algorithms parallel algorithms financial data processing maximum likelihood clustering |
| url | https://www.sajs.co.za/article/view/4149 |
| work_keys_str_mv | AT dieterhendricks highspeeddetectionofemergentmarketclusteringviaanunsupervisedparallelgeneticalgorithm AT timgebbie highspeeddetectionofemergentmarketclusteringviaanunsupervisedparallelgeneticalgorithm AT dianewilcox highspeeddetectionofemergentmarketclusteringviaanunsupervisedparallelgeneticalgorithm |