Foto del docente

Marco Bianchetti

Adjunct professor

Department of Statistical Sciences "Paolo Fortunati"

Research

Keywords: pricing market risk Interest rates Negative rates Quasi Monte Carlo Prudent Valuation Model Risk Valuation Adjustments (CVA/DVA/FVA/etc.) Financial Bubbles Sentiment analysis Portfolio optimization Machine learning

The list below outlines recent research topics and possible M.Sc. thesis on the subject. An associated stage in Intesa Sanpaolo will be considered upon request and availability. Ph.D. thesis can also be considered.

Suggestions for M.Sc. students willing to undergo a research thesis are given at the bottom of this page.

Advices on becoming a quant in finance are available e.g. here and here.

 

A) Machine learning for risk management

Possible research topics are listed below.

  • Application of different machine learning techniques to pricing e risk management of financial instruments.
  • Others to be defined.

Selected references:

  • Previous M.Sc. Thesis and presentations (available upon request)
  • Hernandez, Andres, Model Calibration: Global Optimizer vs. Neural Network (July 3, 2017). Available at SSRN: https://ssrn.com/abstract=2996930.
  • Kondratyev, Alexei, Learning Curve Dynamics with Artificial Neural Networks (April 11, 2018). Available at SSRN: https://ssrn.com/abstract=3041232.
  • Ferguson, Ryan and Green, Andrew David, Deeply Learning Derivatives (October 14, 2018). Available at SSRN: https://ssrn.com/abstract=3244821.
  • Buehler, Hans and Gonon, Lukas and Teichmann, Josef and Wood, Ben and Mohan, Baranidharan and Kochems, Jonathan, Deep Hedging: Hedging Derivatives Under Generic Market Frictions Using Reinforcement Learning (March 19, 2019). Available at SSRN: https://ssrn.com/abstract=3355706
  • Magnus Wiese, Robert Knobloch, Ralf Korn, Peter Kretschmer, Quant GANs: Deep Generation of Financial Time Series, 15 Jul 2019, https://arxiv.org/abs/1907.06673
  • Wiese, Magnus and Bai, Lianjun and Wood, Ben and Buehler, Hans, Deep Hedging: Learning to Simulate Equity Option Markets (October 16, 2019). Available at SSRN: https://ssrn.com/abstract=3470756.
  • Johannes Ruf, Weiguan Wang, Neural networks for option pricing and hedging: a literature review, 13 Nov 2019, https://arxiv.org/abs/1911.05620v1
 

    B) Risk-Oriented portfolio optimization
    Possible research items are listed below.

    • Application of different optimization metaheuristics to solve the optimization problem.
    • Study and implementation of a continuous constrained mapping of the discrete optimization problem.
    • Implementation of different risk and pricing measures to be optimized.
    • Application to complex portfolios, with multiple risk factors, instruments, risk measures and constraints.
    • Formulation of the optimization problem as a Quadratic Unconstrained Binary Optimization ready for D-Wave quantum computer.

    The research will leverage on our previous developments (Matlab code available) and experience (see refs. below).

    Selected references:

    • Previous M.Sc. Thesis and conference presentations (available upon request)
    • S. Kucherenko, Y. Sytsko, "Application of deterministic low-discrepancy sequences to nonlinear global optimization problems". Comp. Optimization and Applications, 30, 3, 287-318 (2005).
    • A. Kondratyev, and G. Giorgidze, "Evolutionary Algos for Optimising MVA", Risk, Dec. 2017. Also available in SSRN, January 3, 2017, https://ssrn.com/abstract=2921822.
    • G. Rosenberg et al., Solving the Optimal Trading Trajectory Problem Using a Quantum Annealer, IEEE Journal of Selected Topics in Signal Processing (JSTSP), Volume 10, Issue 6, 2016, and Proc. of the 8th Workshop on High Performance Computational Finance (WHPCF), p. 7, ACM, 2015. Available in Arxiv https://arxiv.org/abs/1508.06182.

    C) Financial Bubbles

    Possible research items are listed below.

    • Further extensions and improvements to JLS and PSY models for bubble analysis (see refs. below).
    • Study and implementation of further models for bubble analysis, e.g. Jarrow et al. (see ref. below).
    • identification of relevant case studies and application of different models for bubble analysis, development and implementation of bubble indicators, extensive analyses and regular reporting, along the lines of the Financial Crisis Observatory.
    • Extension of the analysis to include sentiment data (see refs. below).

    The research will leverage on our previous developments (Matlab code available) and experience (see refs. below).

    Selected references:

    • Previous M.Sc. Thesis and presentations (available upon request)
    • M. Bianchetti et al., "Are Cryptocurrencies Real Financial Bubbles? Evidence from Quantitative Analyses" (February 24, 2018). Available at SSRN: https://ssrn.com/abstract=3092427.
    • Bianchetti et al, "Brexit or Bremain? Evidence from Bubble Analysis". Published on CEUR-WS: 29-Dec-2016, Proceedings of the 1st Workshop on MIning DAta for financial applicationS (MIDAS 2016), Riva del Garda, Italy, September 19-23, 2016. Edited by: I.Bordino, G. Caldarelli, F. Fumarola, F. Gullo, T. Squartini. Also published in Risk Magazine, 22 June 2016 and in SSRN https://ssrn.com/abstract=2798434.
    • [JLS] V. Filimonov, G. Demos, D. Sornette, "Modified Profile Likelihood Inference and Interval Forecast of the Burst of Financial Bubbles", February 26, 2016, available at SSRN http://ssrn.com/abstract=2739832, and references therein.
    • [PSY] Phillips, Peter C. B. and Shi, Shuping, Financial Bubble Implosion (August 26, 2014). Available at SSRN: https://ssrn.com/abstract=2487601
    • Robert A. Jarrow, "Asset Price Bubbles", Annual Review of Financial Economics, Vol. 7, pp. 201-218, 2015, available at SSRN http://ssrn.com/abstract=2702329
    • Past M.Sc. thesis (A. Salvatori, M. Scaringi, F. Reggiani, C. Del Rio).
    • Sentiment analysis: Richard L. Peterson, "Trading on Sentiment: The Power of Minds Over Markets", Wiley Finance, March 21, 2016.
    • Thomson Reuters Market psych Indices

     

    D) Applications of Quasi Monte Carlo (QMC) methods in finance

    Possible research items are listed below.

    • QMC performance for multi-asset derivatives.
    • Nested QMC performance.
    • QMC performance for different underlying asset dynamics (e.g. Black-Scholes vs Heston vs Local Volatility or jumps).
    • QMC performance for different discretization techniques (e.g. Euler vs Brownian Bridge).
    • Application of global sensitivity analysis techniques.

    The research will leverage on our previous developments (Matlab code available) and experience (see refs. below).

    Selected references:

    • M. Bianchetti, S. Kucherenko, S. Scoleri, “Pricing and Risk Management With High-Dimensional Quasi Monte Carlo and Global Sensitivity Analysis”, Wilmott, 2015: 46–70, July 2015. Available at SSRN at http://ssrn.com/abstract=2592753.
    • StefanoScoleri and Bianchetti, Marco and Kucherenko, Sergei, "Application of Quasi Monte Carlo and Global Sensitivity Analysis to Option Pricing and Greeks" (February 5, 2017). Available at SSRN: https://ssrn.com/abstract=2911698.
    • Past M.Sc. thesis (N. Sedini, A. Tognon, available upon request).

     

    E) Prudent valuation
    Possible research items are listed below.

    • Valuation of opaque financial instruments and their impact on Additional Valuation Adjustments (AVAs).
    • Model risk management.
    • P&L variance test: reduction of risk factors for Market Price Uncertainty and Close-out Cost AVAs, identification of driver risk factors, application of principal component and sensitivity analysis techniques.

    The research will leverage on our previous developments (Matlab code available) and experience (see refs. below).

    Selected references:

     

    F) Modeling financial derivatives
    Possible research items are listed below.

    • Modeling valuation adjustments: Funding Valuation Adjustment (FVA), Margin Valuation Adjustment (MVA), Capital Valuation Adjustment (KVA).
    • Impact of exotic collateral features (e.g. asymmetric collateral, rating downgrade triggers, etc.) on XVAs.
    • Latest SABR model extensions: analysis of recent extensions of the original SABR model, numerical applications to real case studies.
    • Other to be discussed

    The research will leverage on previous developments (Matlab code available) and experience

    Selected references:

     

    Suggestions for M.Sc. students willing to undergo a research thesis

    A research thesis is a short term research project. Suggestions on how to manage a research thesis are available, for example, here (just take inspiration, not each single detail).

    The structure of the thesis should be more or less the following.

    1. Cover page
    2. Abstract (max 1 page)
    3. Introduction: problem definition, state-of-the-art literature, theory/concepts, objectives of the thesis, research questions.
    4. Theoretical background and approach: description of how the problem is theoretically studied.
    5. Applications and results: describe and discuss here in detail the case studies, the results, comparison with the literature, good/weak points, and possible extensions of the work.
    6. Conclusions: answer to objectives\research questions of the thesis, summarize the results and possible directions of future work.
    7. Annexes: detailed theoretical and practical results not addressed in the main chapters, codes used/developed, etc.
    8. References

    Depending on the results, the thesis could lead to a paper to be submitted to a scientific journal.