Deep calibration of financial models: turning theory into practice.

IF 0.7 4区经济学 Q4 BUSINESS, FINANCE

Review of Derivatives Research Pub Date : 2022-01-01 Epub Date: 2021-08-17 DOI:10.1007/s11147-021-09183-7

Patrick Büchel, Michael Kratochwil, Maximilian Nagl, Daniel Rösch

引用次数: 0

Abstract

The calibration of financial models is laborious, time-consuming and expensive, and needs to be performed frequently by financial institutions. Recently, the application of artificial neural networks (ANNs) for model calibration has gained interest. This paper provides the first comprehensive empirical study on the application of ANNs for calibration based on observed market data. We benchmark the performance of the ANN approach against a real-life calibration framework that is in action at a large financial institution. The ANN based calibration framework shows competitive calibration results, roughly four times faster with less computational efforts. Besides speed and efficiency, the resulting model parameters are found to be more stable over time, enabling more reliable risk reports and business decisions. Furthermore, the calibration framework involves multiple validation steps to counteract regulatory concerns regarding its practical application.

查看原文本刊更多论文

金融模型的深度校准：将理论转化为实践

金融模型的校准是费力、耗时和昂贵的，需要金融机构经常执行。近年来，人工神经网络（ANNs）在模型标定中的应用引起了人们的关注。本文首次对基于观察到的市场数据的人工神经网络校准应用进行了全面的实证研究。我们将人工神经网络方法的性能与一家大型金融机构正在运行的现实校准框架进行基准测试。基于人工神经网络的校准框架显示出有竞争力的校准结果，大约快了四倍，计算量更少。除了速度和效率之外，结果模型参数随着时间的推移更加稳定，从而实现更可靠的风险报告和业务决策。此外，校准框架涉及多个验证步骤，以抵消有关其实际应用的监管问题。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Review of Derivatives Research Multiple-

CiteScore

1.40

自引率

0.00%

发文量

期刊介绍： The proliferation of derivative assets during the past two decades is unprecedented. With this growth in derivatives comes the need for financial institutions, institutional investors, and corporations to use sophisticated quantitative techniques to take full advantage of the spectrum of these new financial instruments. Academic research has significantly contributed to our understanding of derivative assets and markets. The growth of derivative asset markets has been accompanied by a commensurate growth in the volume of scientific research. The Review of Derivatives Research provides an international forum for researchers involved in the general areas of derivative assets. The Review publishes high-quality articles dealing with the pricing and hedging of derivative assets on any underlying asset (commodity, interest rate, currency, equity, real estate, traded or non-traded, etc.). Specific topics include but are not limited to: econometric analyses of derivative markets (efficiency, anomalies, performance, etc.) analysis of swap markets market microstructure and volatility issues regulatory and taxation issues credit risk new areas of applications such as corporate finance (capital budgeting, debt innovations), international trade (tariffs and quotas), banking and insurance (embedded options, asset-liability management) risk-sharing issues and the design of optimal derivative securities risk management, management and control valuation and analysis of the options embedded in capital projects valuation and hedging of exotic options new areas for further development (i.e. natural resources, environmental economics. The Review has a double-blind refereeing process. In contrast to the delays in the decision making and publication processes of many current journals, the Review will provide authors with an initial decision within nine weeks of receipt of the manuscript and a goal of publication within six months after acceptance. Finally, a section of the journal is available for rapid publication on `hot'' issues in the market, small technical pieces, and timely essays related to pending legislation and policy. Officially cited as: Rev Deriv Res