Machine Learning (ML) and Artificial Neural Networks (ANN) in Computational Finance

Machine Learning (ML) and Artificial Neural Networks (ANN) in Computational Finance


Abstract

This is one of the first MSc theses to address the full software lifecycle of the analysis (maths), design (Structured Analysis/top-down decomposition) and implementation (C++, Python, ANN, Keras, TensorFlow) to computing option prices and implied volatility under rough Heston model. This new model resolves a number of issues surrounding the original Heston model.

We compare the solutions based on ANNs with more traditional computational solutions; based on our level playing field analysis (that is, we compare “apples with apples”), for this problem the performance of the ANN solution is 7 times slower for option pricing and 17 times slower for implied volatility modelling than traditional methods. Of course, this is only one example but it is hard evidence nonetheless.

There are few articles that discuss the application of ANNs to computational finance and the ones that have been published claim outlandish performance improvements (10,000 times faster) or claim that they can solve 100-factor partial differential equations (PDEs) with Deep Learning techniques.

The full text and pdf of thesis can be found here.

https://www.datasim.nl/blogs/29/msc-theses-on-machine-learning-and-computational-finance-2020

If you have queries please don't hesitate to contact me.

Daniel J. Duffy, PhD

Author/trainer/mentor in computational finance: maths (pure, applied, numerical), ODE/PDE/FDM, C++11/C++20, Python, C#, modern software design

3 年
回复
Cetin Karakus

Global Head of Quantitative and Analytical Solutions at BP

4 年

Like the term “level playing field” Are you by any chance inspired by brexit negotiations?

Johan Marin

Executive Director at Daiwa Capital Markets Europe Ltd

4 年

Thank you for sharing this thesis which runs counter to the trend.

Stephen Channell

C#/F#/C++ Architect/Developer

4 年

Factors of 7 and 14 should help tempter the desire to rebrand trading algo as AI

Daniel J. Duffy, PhD

Author/trainer/mentor in computational finance: maths (pure, applied, numerical), ODE/PDE/FDM, C++11/C++20, Python, C#, modern software design

4 年

Of course, the next phase is to see patterns emerging. Opportunity for young, dynamic and international data scientists? Don't rule out Functional Analysis and RKHS! I feel uneasy with Gradient Descent, maybe I worry too much 0. Inside GD lurks a nasty Euler method. 1. Initial guess must be close to real solution (Analyse Numerique 101). 2. No guarantee that GD is applicable in the first place (assumes cost function is smooth). 3. "Vanishing gradient syndrome" https://en.wikipedia.org/wiki/Vanishing ... nt_problem 4. Learning rate parameter... so many to choose from (ad hoc/trial and error process). 5. Use Armijo?and Wolfe to improve convergence. 6. Modify algorithm by adding momentum. 7. Any you have to compute gradient 1) exact, 2) FDM, 3) AD, 4) complex step method. 8. Convergence to local minimum. 9. The method is iterative, so no true reliable quality of service (QOS). 10. It's not very robust (cf. adversarial examples). Try regularization. "Das ist nicht nur nicht richtig; es ist nicht einmal falsch!" Wolfgang Pauli

要查看或添加评论,请登录