Context Based Predictive Information
We propose a new algorithm called the context-based predictive information (CBPI) for estimating the predictive information (PI) between time series, by utilizing a lossy compression algorithm. The advantage of this approach over existing methods resides in the case of sparse predictive information...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2019-06-01
|
| Series: | Entropy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1099-4300/21/7/645 |
| _version_ | 1811302100493139968 |
|---|---|
| author | Yuval Shalev Irad Ben-Gal |
| author_facet | Yuval Shalev Irad Ben-Gal |
| author_sort | Yuval Shalev |
| collection | DOAJ |
| description | We propose a new algorithm called the context-based predictive information (CBPI) for estimating the predictive information (PI) between time series, by utilizing a lossy compression algorithm. The advantage of this approach over existing methods resides in the case of sparse predictive information (SPI) conditions, where the ratio between the number of informative sequences to uninformative sequences is small. It is shown that the CBPI achieves a better PI estimation than benchmark methods by ignoring uninformative sequences while improving explainability by identifying the informative sequences. We also provide an implementation of the CBPI algorithm on a real dataset of large banks’ stock prices in the U.S. In the last part of this paper, we show how the CBPI algorithm is related to the well-known information bottleneck in its deterministic version. |
| first_indexed | 2024-04-13T07:20:49Z |
| format | Article |
| id | doaj.art-fe864c5e408d4c849bb9ee990d749b9d |
| institution | Directory Open Access Journal |
| issn | 1099-4300 |
| language | English |
| last_indexed | 2024-04-13T07:20:49Z |
| publishDate | 2019-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Entropy |
| spelling | doaj.art-fe864c5e408d4c849bb9ee990d749b9d2022-12-22T02:56:37ZengMDPI AGEntropy1099-43002019-06-0121764510.3390/e21070645e21070645Context Based Predictive InformationYuval Shalev0Irad Ben-Gal1Laboratory for AI, Machine Learning, Business & Data Analytics, Department of Industrial Engineering, The Tel-Aviv University, Ramat-Aviv 6997801, IsraelLaboratory for AI, Machine Learning, Business & Data Analytics, Department of Industrial Engineering, The Tel-Aviv University, Ramat-Aviv 6997801, IsraelWe propose a new algorithm called the context-based predictive information (CBPI) for estimating the predictive information (PI) between time series, by utilizing a lossy compression algorithm. The advantage of this approach over existing methods resides in the case of sparse predictive information (SPI) conditions, where the ratio between the number of informative sequences to uninformative sequences is small. It is shown that the CBPI achieves a better PI estimation than benchmark methods by ignoring uninformative sequences while improving explainability by identifying the informative sequences. We also provide an implementation of the CBPI algorithm on a real dataset of large banks’ stock prices in the U.S. In the last part of this paper, we show how the CBPI algorithm is related to the well-known information bottleneck in its deterministic version.https://www.mdpi.com/1099-4300/21/7/645context treepredictive informationtime series analysisinformation bottleneck |
| spellingShingle | Yuval Shalev Irad Ben-Gal Context Based Predictive Information Entropy context tree predictive information time series analysis information bottleneck |
| title | Context Based Predictive Information |
| title_full | Context Based Predictive Information |
| title_fullStr | Context Based Predictive Information |
| title_full_unstemmed | Context Based Predictive Information |
| title_short | Context Based Predictive Information |
| title_sort | context based predictive information |
| topic | context tree predictive information time series analysis information bottleneck |
| url | https://www.mdpi.com/1099-4300/21/7/645 |
| work_keys_str_mv | AT yuvalshalev contextbasedpredictiveinformation AT iradbengal contextbasedpredictiveinformation |