A neural network-based framework for financial model calibration (2024)

Abstract

A data-driven approach called CaNN (Calibration Neural Network) is proposed to calibrate financial asset price models using an Artificial Neural Network (ANN). Determining optimal values of the model parameters is formulated as training hidden neurons within a machine learning framework, based on available financial option prices. The framework consists of two parts: a forward pass in which we train the weights of the ANN off-line, valuing options under many different asset model parameter settings; and a backward pass, in which we evaluate the trained ANN-solver on-line, aiming to find the weights of the neurons in the input layer. The rapid on-line learning of implied volatility by ANNs, in combination with the use of an adapted parallel global optimization method, tackles the computation bottleneck and provides a fast and reliable technique for calibrating model parameters while avoiding, as much as possible, getting stuck in local minima. Numerical experiments confirm that this machine-learning framework can be employed to calibrate parameters of high-dimensional stochastic volatility models efficiently and accurately.

Original languageEnglish
Article number9
Pages (from-to)1-28
Number of pages28
JournalMathematics in Industry
Volume9
Issue number1
DOIs
Publication statusPublished - 2019

Keywords

  • Artificial neural networks
  • Asset pricing model
  • Computational finance
  • Global optimization
  • Machine learning
  • Model calibration
  • Parallel computing

Access to Document

  • s13362-019-0066-7Final published version, 2.24 MBLicence: CC BY

Other files and links

    Fingerprint

    Dive into the research topics of 'A neural network-based framework for financial model calibration'. Together they form a unique fingerprint.

    View full fingerprint

    Cite this

    • APA
    • Author
    • BIBTEX
    • Harvard
    • Standard
    • RIS
    • Vancouver

    Liu, S., Borovykh, A., Grzelak, L. (2019). A neural network-based framework for financial model calibration. Mathematics in Industry, 9(1), 1-28. Article 9. https://doi.org/10.1186/s13362-019-0066-7

    Liu, Shuaiqiang ; Borovykh, Anastasia ; Grzelak, Lech et al. / A neural network-based framework for financial model calibration. In: Mathematics in Industry. 2019 ; Vol. 9, No. 1. pp. 1-28.

    @article{f8d68034b70d431cb905d5b47292749f,

    title = "A neural network-based framework for financial model calibration",

    abstract = "A data-driven approach called CaNN (Calibration Neural Network) is proposed to calibrate financial asset price models using an Artificial Neural Network (ANN). Determining optimal values of the model parameters is formulated as training hidden neurons within a machine learning framework, based on available financial option prices. The framework consists of two parts: a forward pass in which we train the weights of the ANN off-line, valuing options under many different asset model parameter settings; and a backward pass, in which we evaluate the trained ANN-solver on-line, aiming to find the weights of the neurons in the input layer. The rapid on-line learning of implied volatility by ANNs, in combination with the use of an adapted parallel global optimization method, tackles the computation bottleneck and provides a fast and reliable technique for calibrating model parameters while avoiding, as much as possible, getting stuck in local minima. Numerical experiments confirm that this machine-learning framework can be employed to calibrate parameters of high-dimensional stochastic volatility models efficiently and accurately.",

    keywords = "Artificial neural networks, Asset pricing model, Computational finance, Global optimization, Machine learning, Model calibration, Parallel computing",

    author = "Shuaiqiang Liu and Anastasia Borovykh and Lech Grzelak and Oosterlee, {Cornelis W.}",

    year = "2019",

    doi = "10.1186/s13362-019-0066-7",

    language = "English",

    volume = "9",

    pages = "1--28",

    journal = "Mathematics in Industry",

    issn = "1612-3956",

    publisher = "SpringerOpen",

    number = "1",

    }

    Liu, S, Borovykh, A, Grzelak, L 2019, 'A neural network-based framework for financial model calibration', Mathematics in Industry, vol. 9, no. 1, 9, pp. 1-28. https://doi.org/10.1186/s13362-019-0066-7

    A neural network-based framework for financial model calibration. / Liu, Shuaiqiang; Borovykh, Anastasia; Grzelak, Lech et al.
    In: Mathematics in Industry, Vol. 9, No. 1, 9, 2019, p. 1-28.

    Research output: Contribution to journalArticleScientificpeer-review

    TY - JOUR

    T1 - A neural network-based framework for financial model calibration

    AU - Liu, Shuaiqiang

    AU - Borovykh, Anastasia

    AU - Grzelak, Lech

    AU - Oosterlee, Cornelis W.

    PY - 2019

    Y1 - 2019

    N2 - A data-driven approach called CaNN (Calibration Neural Network) is proposed to calibrate financial asset price models using an Artificial Neural Network (ANN). Determining optimal values of the model parameters is formulated as training hidden neurons within a machine learning framework, based on available financial option prices. The framework consists of two parts: a forward pass in which we train the weights of the ANN off-line, valuing options under many different asset model parameter settings; and a backward pass, in which we evaluate the trained ANN-solver on-line, aiming to find the weights of the neurons in the input layer. The rapid on-line learning of implied volatility by ANNs, in combination with the use of an adapted parallel global optimization method, tackles the computation bottleneck and provides a fast and reliable technique for calibrating model parameters while avoiding, as much as possible, getting stuck in local minima. Numerical experiments confirm that this machine-learning framework can be employed to calibrate parameters of high-dimensional stochastic volatility models efficiently and accurately.

    AB - A data-driven approach called CaNN (Calibration Neural Network) is proposed to calibrate financial asset price models using an Artificial Neural Network (ANN). Determining optimal values of the model parameters is formulated as training hidden neurons within a machine learning framework, based on available financial option prices. The framework consists of two parts: a forward pass in which we train the weights of the ANN off-line, valuing options under many different asset model parameter settings; and a backward pass, in which we evaluate the trained ANN-solver on-line, aiming to find the weights of the neurons in the input layer. The rapid on-line learning of implied volatility by ANNs, in combination with the use of an adapted parallel global optimization method, tackles the computation bottleneck and provides a fast and reliable technique for calibrating model parameters while avoiding, as much as possible, getting stuck in local minima. Numerical experiments confirm that this machine-learning framework can be employed to calibrate parameters of high-dimensional stochastic volatility models efficiently and accurately.

    KW - Artificial neural networks

    KW - Asset pricing model

    KW - Computational finance

    KW - Global optimization

    KW - Machine learning

    KW - Model calibration

    KW - Parallel computing

    UR - http://www.scopus.com/inward/record.url?scp=85071941149&partnerID=8YFLogxK

    U2 - 10.1186/s13362-019-0066-7

    DO - 10.1186/s13362-019-0066-7

    M3 - Article

    AN - SCOPUS:85071941149

    SN - 1612-3956

    VL - 9

    SP - 1

    EP - 28

    JO - Mathematics in Industry

    JF - Mathematics in Industry

    IS - 1

    M1 - 9

    ER -

    Liu S, Borovykh A, Grzelak L, Oosterlee CW. A neural network-based framework for financial model calibration. Mathematics in Industry. 2019;9(1):1-28. 9. doi: 10.1186/s13362-019-0066-7

    A neural network-based framework for financial model calibration (2024)
    Top Articles
    Latest Posts
    Article information

    Author: Horacio Brakus JD

    Last Updated:

    Views: 6467

    Rating: 4 / 5 (51 voted)

    Reviews: 90% of readers found this page helpful

    Author information

    Name: Horacio Brakus JD

    Birthday: 1999-08-21

    Address: Apt. 524 43384 Minnie Prairie, South Edda, MA 62804

    Phone: +5931039998219

    Job: Sales Strategist

    Hobby: Sculling, Kitesurfing, Orienteering, Painting, Computer programming, Creative writing, Scuba diving

    Introduction: My name is Horacio Brakus JD, I am a lively, splendid, jolly, vivacious, vast, cheerful, agreeable person who loves writing and wants to share my knowledge and understanding with you.