tsGT: Stochastic Time Series Modeling With Transformer
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916190839898112 |
|---|---|
| author | Kuciński, Łukasz Drzewakowski, Witold Olko, Mateusz Kozakowski, Piotr Maziarka, Łukasz Nowakowska, Marta Emilia Kaiser, Łukasz Miłoś, Piotr |
| author_facet | Kuciński, Łukasz Drzewakowski, Witold Olko, Mateusz Kozakowski, Piotr Maziarka, Łukasz Nowakowska, Marta Emilia Kaiser, Łukasz Miłoś, Piotr |
| contents | Time series methods are of fundamental importance in virtually any field of science that deals with temporally structured data. Recently, there has been a surge of deterministic transformer models with time series-specific architectural biases. In this paper, we go in a different direction by introducing tsGT, a stochastic time series model built on a general-purpose transformer architecture. We focus on using a well-known and theoretically justified rolling window backtesting and evaluation protocol. We show that tsGT outperforms the state-of-the-art models on MAD and RMSE, and surpasses its stochastic peers on QL and CRPS, on four commonly used datasets. We complement these results with a detailed analysis of tsGT's ability to model the data distribution and predict marginal quantile values. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_05713 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | tsGT: Stochastic Time Series Modeling With Transformer Kuciński, Łukasz Drzewakowski, Witold Olko, Mateusz Kozakowski, Piotr Maziarka, Łukasz Nowakowska, Marta Emilia Kaiser, Łukasz Miłoś, Piotr Machine Learning Time series methods are of fundamental importance in virtually any field of science that deals with temporally structured data. Recently, there has been a surge of deterministic transformer models with time series-specific architectural biases. In this paper, we go in a different direction by introducing tsGT, a stochastic time series model built on a general-purpose transformer architecture. We focus on using a well-known and theoretically justified rolling window backtesting and evaluation protocol. We show that tsGT outperforms the state-of-the-art models on MAD and RMSE, and surpasses its stochastic peers on QL and CRPS, on four commonly used datasets. We complement these results with a detailed analysis of tsGT's ability to model the data distribution and predict marginal quantile values. |
| title | tsGT: Stochastic Time Series Modeling With Transformer |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2403.05713 |