Multivariate Segment Expandable Encoder-Decoder Model for Time Series Forecasting
Accurate time series forecasting is critical in a variety of fields, including transportation, weather prediction, energy management, infrastructure monitoring, and finance. Forecasting highly skewed and heavy-tailed time series, particularly in multivariate environments, is still difficult. In thes...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10781401/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841533429098741760 |
|---|---|
| author | Yanhong Li David C. Anastasiu |
| author_facet | Yanhong Li David C. Anastasiu |
| author_sort | Yanhong Li |
| collection | DOAJ |
| description | Accurate time series forecasting is critical in a variety of fields, including transportation, weather prediction, energy management, infrastructure monitoring, and finance. Forecasting highly skewed and heavy-tailed time series, particularly in multivariate environments, is still difficult. In these cases, accurately capturing the relationships between variables is critical for successful model design. This is especially true when dealing with extreme events like droughts or floods in streamflow forecasting, which can have severe consequences on public safety and social well-being. We present the Multivariate Segment-Expandable Encoder Decoder (MSEED), a novel framework designed to address the challenges of extreme-adaptive multivariate time series forecasting. MSEED features a hierarchical encoder-decoder architecture, a short-term-enhanced subnet, and a feature assembling layer that integrates spatial and temporal information across multivariate inputs. By capturing quantile distributions across segmented subsequences at multiple scales, the model is able to detect complex patterns, enhancing both the accuracy and robustness of forecasts. Additionally, MSEED incorporates a simple vanilla encoder-decoder model for strengthening rolling predictions. The framework has been tested on four challenging real-world datasets, focusing on two critical forecasting scenarios: long-term predictions (three days ahead) and rolling predictions (every four hours) to simulate real-time decision-making in water resource management. MSEED consistently outperforms state-of-the-art models, showing improvements in forecasting accuracy ranging from 18% to 74%. |
| format | Article |
| id | doaj-art-a2f299c4e5b74b02b888a980680d249f |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-a2f299c4e5b74b02b888a980680d249f2025-01-16T00:02:06ZengIEEEIEEE Access2169-35362024-01-011218501218502610.1109/ACCESS.2024.351325610781401Multivariate Segment Expandable Encoder-Decoder Model for Time Series ForecastingYanhong Li0https://orcid.org/0009-0008-8300-9516David C. Anastasiu1https://orcid.org/0000-0002-8604-9248Computer Science and Engineering Department, Santa Clara University, Santa Clara, CA, USAComputer Science and Engineering Department, Santa Clara University, Santa Clara, CA, USAAccurate time series forecasting is critical in a variety of fields, including transportation, weather prediction, energy management, infrastructure monitoring, and finance. Forecasting highly skewed and heavy-tailed time series, particularly in multivariate environments, is still difficult. In these cases, accurately capturing the relationships between variables is critical for successful model design. This is especially true when dealing with extreme events like droughts or floods in streamflow forecasting, which can have severe consequences on public safety and social well-being. We present the Multivariate Segment-Expandable Encoder Decoder (MSEED), a novel framework designed to address the challenges of extreme-adaptive multivariate time series forecasting. MSEED features a hierarchical encoder-decoder architecture, a short-term-enhanced subnet, and a feature assembling layer that integrates spatial and temporal information across multivariate inputs. By capturing quantile distributions across segmented subsequences at multiple scales, the model is able to detect complex patterns, enhancing both the accuracy and robustness of forecasts. Additionally, MSEED incorporates a simple vanilla encoder-decoder model for strengthening rolling predictions. The framework has been tested on four challenging real-world datasets, focusing on two critical forecasting scenarios: long-term predictions (three days ahead) and rolling predictions (every four hours) to simulate real-time decision-making in water resource management. MSEED consistently outperforms state-of-the-art models, showing improvements in forecasting accuracy ranging from 18% to 74%.https://ieeexplore.ieee.org/document/10781401/Deep learningrepresentation learningoversampling policystreamflow predictionhydrologic predictionLSTM |
| spellingShingle | Yanhong Li David C. Anastasiu Multivariate Segment Expandable Encoder-Decoder Model for Time Series Forecasting IEEE Access Deep learning representation learning oversampling policy streamflow prediction hydrologic prediction LSTM |
| title | Multivariate Segment Expandable Encoder-Decoder Model for Time Series Forecasting |
| title_full | Multivariate Segment Expandable Encoder-Decoder Model for Time Series Forecasting |
| title_fullStr | Multivariate Segment Expandable Encoder-Decoder Model for Time Series Forecasting |
| title_full_unstemmed | Multivariate Segment Expandable Encoder-Decoder Model for Time Series Forecasting |
| title_short | Multivariate Segment Expandable Encoder-Decoder Model for Time Series Forecasting |
| title_sort | multivariate segment expandable encoder decoder model for time series forecasting |
| topic | Deep learning representation learning oversampling policy streamflow prediction hydrologic prediction LSTM |
| url | https://ieeexplore.ieee.org/document/10781401/ |
| work_keys_str_mv | AT yanhongli multivariatesegmentexpandableencoderdecodermodelfortimeseriesforecasting AT davidcanastasiu multivariatesegmentexpandableencoderdecodermodelfortimeseriesforecasting |