Luca Capriotti (EMEA head of Quantitative Strategies Global Credit Products, Credit Suisse) will give a seminar at the Scuola Normale Superiore, Pisa on Thursday 19 June 2014, 13:00 Aula Bianchi.
The title and abstract are
Ulisse Dini, Adjoint Algorithmic Differentiation and the Efficient Risk Management of Credit Portfolios
Adjoint Algorithmic Differentiation (AAD) is one of the principal innovations in risk management of the recent times. In this talk I will show how AAD can be used to risk manage in real time credit products, both exotic and vanilla, whether valued with Monte Carlo or with semi-analytical methods. I will show how by combining adjoint ideas with the implicit function (Dini's) theorem one can avoid repeating multiple times the calibration of the hazard rate curves which, especially for flow products, often represent the bottle neck in the computation of spread and interest rate risk. This typically results in orders of magnitudes savings in computation time with respect to standard methods. The adjoint of the calibration step can be naturally combined with the adjoint of the pricing step. This allows one to compute the risk of portfolios of credit products faster than computing the portfolio value alone, thus making risk management in real time possible.
Moreover on Friday 20 June 2014 he will give a mini-course on
Real Time Risk Management with Adjoint Algorithmic Differentiaton
The agenda is attached below. Interested people are invited to participate.
For more information, contact Stefano Marmi (stefano.marmi(a)sns.it), Fabrizio Lillo (fabrizio.lillo(a)sns.it), or Giacomo Bormetti (giacomo.bormetti(a)sns.it).
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
Friday June 20, Aula Bianchi, 11.00-12.30
Friday June 20, Aula Bianchi, 14.00-15.30
Adjoint Algorithmic Differentiation (AAD) is one of the principal innovations in risk management of the recent times. In this minicourse I will introduce AAD and show how it can be used to implement the calculation of price sensitivities in complete generality and with minimal analytical effort. The focus will be the application to Monte Carlo methods - generally the most challenging from the computational point of view. With several examples I will illustrate the workings of AAD and demonstrate how it can be straightforwardly implemented to reduce the computation time of the risk of any portfolio by order of magnitudes.
1. Pathwise Derivative Method
2. Algebraic Adjoint Approaches
3. Adjoint Algorithmic Differentiation (AAD)
4. AAD as a Design Paradigm
5. AAD and the Pathwise Derivative Method
6. First Applications
7. Case Study: Adjoint Greeks for the Libor Market Model
8. Correlation Risk and Binning Techniques
9. Case Study: Correlation Greeks for Basket Default Contracts