Autoencoder-Enhanced Clustering: A Dimensionality Reduction Approach to Financial Time Series
While Machine Learning significantly boosts the performance of predictive models, its efficacy varies across different data dimensions. It is essential to cluster time series data of similar characteristics, particularly in the financial sector. However, clustering financial time series data poses c...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10415421/ |
| _version_ | 1827357555236536320 |
|---|---|
| author | Daniel Gonzalez Cortes Enrique Onieva Iker Pastor Lopez Laura Trinchera Jian Wu |
| author_facet | Daniel Gonzalez Cortes Enrique Onieva Iker Pastor Lopez Laura Trinchera Jian Wu |
| author_sort | Daniel Gonzalez Cortes |
| collection | DOAJ |
| description | While Machine Learning significantly boosts the performance of predictive models, its efficacy varies across different data dimensions. It is essential to cluster time series data of similar characteristics, particularly in the financial sector. However, clustering financial time series data poses considerable challenges due to the market’s inherent complexity and multidimensionality. To address these issues, our study introduces a novel clustering framework that leverages autoencoders for a compressed yet informative representation of financial time series. We rigorously evaluate our approach through multiple dimensionality reduction and clustering algorithms, applying it to key financial indices, including IBEX-35, CAC-40, DAX-30, S&P 500, and FTSE 100. Our findings consistently demonstrate that incorporating autoencoders significantly enhances the granularity and quality of clustering, effectively isolating distinct categories of financial time series. Our findings carry significant ramifications for the financial industry. By refining clustering methodologies, we set the stage for increasingly accurate financial predictive models, offering valuable insights for optimizing investment strategies and enhancing risk management. |
| first_indexed | 2024-03-08T05:34:34Z |
| format | Article |
| id | doaj.art-04c239e64151499aac3ed9a8a80092f9 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-03-08T05:34:34Z |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-04c239e64151499aac3ed9a8a80092f92024-02-06T00:01:10ZengIEEEIEEE Access2169-35362024-01-0112169991700910.1109/ACCESS.2024.335941310415421Autoencoder-Enhanced Clustering: A Dimensionality Reduction Approach to Financial Time SeriesDaniel Gonzalez Cortes0https://orcid.org/0000-0002-5170-9883Enrique Onieva1https://orcid.org/0000-0001-9581-1823Iker Pastor Lopez2https://orcid.org/0000-0002-3068-6248Laura Trinchera3https://orcid.org/0000-0001-9679-0956Jian Wu4https://orcid.org/0000-0002-0855-1881NEOMA Business School, Mont-Saint-Aignan, FranceFaculty of Engineering, University of Deusto, Bilbao, SpainFaculty of Engineering, University of Deusto, Bilbao, SpainNEOMA Business School, Mont-Saint-Aignan, FranceNEOMA Business School, Mont-Saint-Aignan, FranceWhile Machine Learning significantly boosts the performance of predictive models, its efficacy varies across different data dimensions. It is essential to cluster time series data of similar characteristics, particularly in the financial sector. However, clustering financial time series data poses considerable challenges due to the market’s inherent complexity and multidimensionality. To address these issues, our study introduces a novel clustering framework that leverages autoencoders for a compressed yet informative representation of financial time series. We rigorously evaluate our approach through multiple dimensionality reduction and clustering algorithms, applying it to key financial indices, including IBEX-35, CAC-40, DAX-30, S&P 500, and FTSE 100. Our findings consistently demonstrate that incorporating autoencoders significantly enhances the granularity and quality of clustering, effectively isolating distinct categories of financial time series. Our findings carry significant ramifications for the financial industry. By refining clustering methodologies, we set the stage for increasingly accurate financial predictive models, offering valuable insights for optimizing investment strategies and enhancing risk management.https://ieeexplore.ieee.org/document/10415421/Clustering methodsdata compressionfinancial data processingneural network applicationstime series |
| spellingShingle | Daniel Gonzalez Cortes Enrique Onieva Iker Pastor Lopez Laura Trinchera Jian Wu Autoencoder-Enhanced Clustering: A Dimensionality Reduction Approach to Financial Time Series IEEE Access Clustering methods data compression financial data processing neural network applications time series |
| title | Autoencoder-Enhanced Clustering: A Dimensionality Reduction Approach to Financial Time Series |
| title_full | Autoencoder-Enhanced Clustering: A Dimensionality Reduction Approach to Financial Time Series |
| title_fullStr | Autoencoder-Enhanced Clustering: A Dimensionality Reduction Approach to Financial Time Series |
| title_full_unstemmed | Autoencoder-Enhanced Clustering: A Dimensionality Reduction Approach to Financial Time Series |
| title_short | Autoencoder-Enhanced Clustering: A Dimensionality Reduction Approach to Financial Time Series |
| title_sort | autoencoder enhanced clustering a dimensionality reduction approach to financial time series |
| topic | Clustering methods data compression financial data processing neural network applications time series |
| url | https://ieeexplore.ieee.org/document/10415421/ |
| work_keys_str_mv | AT danielgonzalezcortes autoencoderenhancedclusteringadimensionalityreductionapproachtofinancialtimeseries AT enriqueonieva autoencoderenhancedclusteringadimensionalityreductionapproachtofinancialtimeseries AT ikerpastorlopez autoencoderenhancedclusteringadimensionalityreductionapproachtofinancialtimeseries AT lauratrinchera autoencoderenhancedclusteringadimensionalityreductionapproachtofinancialtimeseries AT jianwu autoencoderenhancedclusteringadimensionalityreductionapproachtofinancialtimeseries |