Time-Series Foundation AI Model for Value-at-Risk Forecasting

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Goel, Anubha, Pasricha, Puneet, Kanniainen, Juho
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866918016378208256
author Goel, Anubha
Pasricha, Puneet
Kanniainen, Juho
author_facet Goel, Anubha
Pasricha, Puneet
Kanniainen, Juho
contents This study is the first to analyze the performance of a time-series foundation AI model for Value-at-Risk (VaR), which essentially forecasts the left-tail quantiles of returns. Foundation models, pre-trained on diverse datasets, can be applied in a zero-shot setting with minimal data or further improved through finetuning. We compare Google's TimesFM model to conventional parametric and non-parametric models, including GARCH and Generalized Autoregressive Score (GAS), using 19 years of daily returns from the SP 100 index and its constituents. Backtesting with over 8.5 years of out-of-sample data shows that the fine-tuned foundation model consistently outperforms traditional methods in actual-over-expected ratios. For the quantile score loss function, it performs comparably to the best econometric model, GAS. Overall, the foundation model ranks as the best or among the top performers across the 0.01, 0.025, 0.05, and 0.1 quantile forecasting. Fine-tuning significantly improves accuracy, showing that zero-shot use is not optimal for VaR.
format Preprint
id arxiv_https___arxiv_org_abs_2410_11773
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Time-Series Foundation AI Model for Value-at-Risk Forecasting
Goel, Anubha
Pasricha, Puneet
Kanniainen, Juho
Risk Management
Artificial Intelligence
This study is the first to analyze the performance of a time-series foundation AI model for Value-at-Risk (VaR), which essentially forecasts the left-tail quantiles of returns. Foundation models, pre-trained on diverse datasets, can be applied in a zero-shot setting with minimal data or further improved through finetuning. We compare Google's TimesFM model to conventional parametric and non-parametric models, including GARCH and Generalized Autoregressive Score (GAS), using 19 years of daily returns from the SP 100 index and its constituents. Backtesting with over 8.5 years of out-of-sample data shows that the fine-tuned foundation model consistently outperforms traditional methods in actual-over-expected ratios. For the quantile score loss function, it performs comparably to the best econometric model, GAS. Overall, the foundation model ranks as the best or among the top performers across the 0.01, 0.025, 0.05, and 0.1 quantile forecasting. Fine-tuning significantly improves accuracy, showing that zero-shot use is not optimal for VaR.
title Time-Series Foundation AI Model for Value-at-Risk Forecasting
topic Risk Management
Artificial Intelligence
url https://arxiv.org/abs/2410.11773