Grant: Discovery Grants Program - Group
Ryerson University
$20,000.00
Modeling multidimensional stochastic systems
EF 2009-10
Résumé
The general goal of this research project is the study of multivariate stochastic models with stochastic correlation and jumps and their quantitative impact in the risk of financial markets and the pricing of exotic products.
Taking into account the joint behavior of different assets in the model would bring a better assessment of such risks and a more accurate understanding of some of the complex derivatives that financial institutions have issued in recent years. Such products have been mispriced in practice by the use of inadequate models. Consequently the under evaluation of risk in highly nonlinear portfolios containing multivariate derivatives is considered one of the main reasons of the current world financial crisis. This revealed to be particularly true in the ongoing crisis of the credit sector, where the incorrect evaluation of Collateral Debt Obligations (CDO's) provoked the bankruptcy of key institutions.
We will examine properties of several novel multivariate processes driven by non Gaussian noises, computing expected values of certain functionals and describing their asymptotic behavior as stationarity, mixing and ergodic properties. Based on the asymptotic properties of these models, together with their Markovian structure, we will implement martingale estimating procedures via conditional characteristic functions (CCF), moments and approximations of the likelihood. We use these properties for calibration and testing of these models to real market data.
We also consider the pricing of complex financial derivatives whose payoffs depend on the collective behavior of several underlying assets assuming the dynamic described by the models above. Basket and Spread Options, Mountain Range Derivatives and Collateralized Debt Obligations are our specific targets. Pricing expressions will be obtained by techniques based on the CCF, the Green function, the method of images and the analytical/numerical solutions of PDEs.
Programme
Discovery Grants Program - Group
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
376331-2009
Name nom
EscobarAnel, MarcosLazaro
Department département
Mathematics
Organization id
43
Institution établissement
Ryerson University
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2009
Competition year année de concours
2009
Award amount
20000
Program id
RGPGP
Program nane en
Discovery Grants Program - Group
Program name fr
Programme de subventions à la découverte - équipe
Group en
Discovery grants programs
Group fr
Programmes de subventions à la découverte
Committee code
337
Committee name en
Pure & Applied Maths - B
Committee name fr
Maths pures et appliquées - B
Area of application code
1205
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Mathematical sciences
Area of application fr
Sciences mathématiques
Research subject code
3007
Research subject group en
Statistics and probability
Research subject group fr
Statistique et probabilité
Research subject en
Stochastic processes
Research subject fr
Processus stochastiques
Partie
1
Nb partie
5
Application title
Modeling multidimensional stochastic systems
Keyword
NA
Application summary
The general goal of this research project is the study of multivariate stochastic models with stochastic correlation and jumps and their quantitative impact in the risk of financial markets and the pricing of exotic products.
Taking into account the joint behavior of different assets in the model would bring a better assessment of such risks and a more accurate understanding of some of the complex derivatives that financial institutions have issued in recent years. Such products have been mispriced in practice by the use of inadequate models. Consequently the under evaluation of risk in highly nonlinear portfolios containing multivariate derivatives is considered one of the main reasons of the current world financial crisis. This revealed to be particularly true in the ongoing crisis of the credit sector, where the incorrect evaluation of Collateral Debt Obligations (CDO's) provoked the bankruptcy of key institutions.
We will examine properties of several novel multivariate processes driven by non Gaussian noises, computing expected values of certain functionals and describing their asymptotic behavior as stationarity, mixing and ergodic properties. Based on the asymptotic properties of these models, together with their Markovian structure, we will implement martingale estimating procedures via conditional characteristic functions (CCF), moments and approximations of the likelihood. We use these properties for calibration and testing of these models to real market data.
We also consider the pricing of complex financial derivatives whose payoffs depend on the collective behavior of several underlying assets assuming the dynamic described by the models above. Basket and Spread Options, Mountain Range Derivatives and Collateralized Debt Obligations are our specific targets. Pricing expressions will be obtained by techniques based on the CCF, the Green function, the method of images and the analytical/numerical solutions of PDEs.