Deep Estimation for Volatility Forecasting

Léo Parent ()
Additional contact information
Léo Parent: UP1 EMS - Université Paris 1 Panthéon-Sorbonne - École de Management de la Sorbonne - UP1 - Université Paris 1 Panthéon-Sorbonne

Working Papers from HAL

Abstract: The use of deep neural networks (DNNs) for the calibration of volatility models applied to pricing and hedging issues has led to abundant academic literature. In contrast, few works utilize these tools for model estimation with a focus on volatility forecasting. Based on this observation, this article introduces an innovative deep estimation method using historical data, specifically designed for volatility forecasting. To illustrate this method, the article focuses on estimating a rough path-dependent volatility (RPDV) model, which is well-suited to the prediction objective and very complex to estimate using standard approaches. After formalizing the estimation problem within the framework of Bayesian decision theory, the article details the methodology for constructing the estimator function. Finally, a comprehensive evaluation of the estimation approach is conducted using both synthetic and market data to assess its performance.

Keywords: PDV model; Volatility forecasting; Tough volatility; Tough path-dependent volatility model; Deep learning; Deep calibration; Bayesian decision theory (search for similar items in EconPapers)
Date: 2024-10-24
Note: View the original document on HAL open archive server: https://hal.science/hal-04751392v1
References: Add references at CitEc
Citations:

Downloads: (external link)
https://hal.science/hal-04751392v1/document (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:hal:wpaper:hal-04751392

Access Statistics for this paper

More papers in Working Papers from HAL
Bibliographic data for series maintained by CCSD ().