LOB-Bench: Benchmarking Generative AI for Finance -- an Application to Limit Order Book Data
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909649682300928 |
|---|---|
| author | Nagy, Peer Frey, Sascha Li, Kang Sarkar, Bidipta Vyetrenko, Svitlana Zohren, Stefan Calinescu, Ani Foerster, Jakob |
| author_facet | Nagy, Peer Frey, Sascha Li, Kang Sarkar, Bidipta Vyetrenko, Svitlana Zohren, Stefan Calinescu, Ani Foerster, Jakob |
| contents | While financial data presents one of the most challenging and interesting sequence modelling tasks due to high noise, heavy tails, and strategic interactions, progress in this area has been hindered by the lack of consensus on quantitative evaluation paradigms. To address this, we present LOB-Bench, a benchmark, implemented in python, designed to evaluate the quality and realism of generative message-by-order data for limit order books (LOB) in the LOBSTER format. Our framework measures distributional differences in conditional and unconditional statistics between generated and real LOB data, supporting flexible multivariate statistical evaluation. The benchmark also includes features commonly used LOB statistics such as spread, order book volumes, order imbalance, and message inter-arrival times, along with scores from a trained discriminator network. Lastly, LOB-Bench contains "market impact metrics", i.e. the cross-correlations and price response functions for specific events in the data. We benchmark generative autoregressive state-space models, a (C)GAN, as well as a parametric LOB model and find that the autoregressive GenAI approach beats traditional model classes. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_09172 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | LOB-Bench: Benchmarking Generative AI for Finance -- an Application to Limit Order Book Data Nagy, Peer Frey, Sascha Li, Kang Sarkar, Bidipta Vyetrenko, Svitlana Zohren, Stefan Calinescu, Ani Foerster, Jakob Machine Learning Computational Engineering, Finance, and Science Computational Finance Trading and Market Microstructure While financial data presents one of the most challenging and interesting sequence modelling tasks due to high noise, heavy tails, and strategic interactions, progress in this area has been hindered by the lack of consensus on quantitative evaluation paradigms. To address this, we present LOB-Bench, a benchmark, implemented in python, designed to evaluate the quality and realism of generative message-by-order data for limit order books (LOB) in the LOBSTER format. Our framework measures distributional differences in conditional and unconditional statistics between generated and real LOB data, supporting flexible multivariate statistical evaluation. The benchmark also includes features commonly used LOB statistics such as spread, order book volumes, order imbalance, and message inter-arrival times, along with scores from a trained discriminator network. Lastly, LOB-Bench contains "market impact metrics", i.e. the cross-correlations and price response functions for specific events in the data. We benchmark generative autoregressive state-space models, a (C)GAN, as well as a parametric LOB model and find that the autoregressive GenAI approach beats traditional model classes. |
| title | LOB-Bench: Benchmarking Generative AI for Finance -- an Application to Limit Order Book Data |
| topic | Machine Learning Computational Engineering, Finance, and Science Computational Finance Trading and Market Microstructure |
| url | https://arxiv.org/abs/2502.09172 |