Pablo Olivares from University of Toronto will be speaking this Wednesday 2PM in DM 409A.
He is being considered for a visiting lecturer position. Title and abstract below:
Title: Non Gaussian Risk Models in Finance and Actuarial Sciences
Abstract:
Most existing stochastic dynamical models are based on the assumption
of the Gaussian distribution of their constituents although some
common observed non-linear phenomena as asymmetry, heavy tails and
stochastic correlation indicate the presence of non Gaussian random
disturbances.
Models driven by noises with richer properties, as Levy processes and
stochastic volatility for example, allowing for random jumps and
larger oscillations have been proposed recently.
In this context we introduce stable ARMA and GARCH, jump-diffusion and
principal component stochastic volatility models in finance, as well
as a spectrally negative general Levy process in insurance. We discuss
their asymptotic behavior and the problem of parameter estimation.
Some applications to Value at Risk calculations and pricing of
multivariate financial derivatives are shown.
In addition we review some experiences in applied projects with
financial companies.