Machine Learning for finance

Lecturers:  J.D. Fermanian and H. Pham
Period:  
Term 3
ECTS: 6
Schedule:  3 hours per week

The aim of this lecture is to introduce some fundamental concepts and techniques from machine learning and deep learning with a view towards important and recent applications in finance : This includes advanced techniques in scoring, text mining for stock market prediction, hedging/pricing of options, calibration of models,  optimal transport and robust finance,  numerical resolution of high-dimensional non-linear partial differential equations arising for instance in stochastic control and portfolio selection, market generators.  

 

Part I: Fundamental concepts from machine learning

       1. Presentation of the main machine learning algorithms 

       2. Overlearning: penalization, regularization, cross-validation

       3. Presentation of the main scoring techniques

       4. Deep learning: Multi-layer feedforward neural networks, convolutional, recurrent networks.  Backpropagation, stochastic gradient for training.                Implementation with  TensorFlow

 

Part II:  Applications in finance

        1. Text data processing and stock market prediction 

        2.  Deep hedging and deep calibration 

        3.  Deep  reinforcement learning and applications

            - Q-learning algorithms, policy gradient, actor-critic algorithm 

            -  Stochastic control and  portfolio optimisation

           -  Nonlinear PDE, American option pricing, counterparty risk (CVA).

      4.   Market generators and deep simulation

 

References

[1] A. Bachouch, C. Huré, N. Langrené, H. Pham : Deep neural networks algorithms for stochastic control problems on finite horizon, part II, numerical applications : to appear in Methodology and Computing in Applied Probability.

 

[2] C. Bayer, B. Horvath, A. Muguruza, B. Stemper, M. Tomas : On deep calibration of (rough) stochastic volatility models, arXiv : 1908.08806

 

[3] D. Bloch : Machine learning : models and algorithms, {\it Quantitative Analytics}, 2020. 

 

[4] H. Buehler, L. Gonon, J. Teichmann, B. Wood : Deep hedging, {\it Quantitative Finance}, 19(8), 1271-1291, 2019. 

 

[5] S. Eckstein and M. Kupper : Computation of optimal transport and related hedging problems  via penalization and neural networks.  Applied Mathematics and Optimization, 1-29, 2019. 

 

[6] C. Huré, H. Pham, X. Warin : Deep backward schemes for high-dimensional nonlinear PDEs,  Mathematics of Computation. 2019

 

[7] P. Henry-Labordère : Generative models for financial data, SSRN 3408007, 2019

 

[8] M. Lopez de Prado :  Advances in machine learning, Wiley, 2016.