Dr. Zdravko Botev
University of New South Wales
The distribution of sums of dependent log-normal variables has applications in, for example, the valuation of an asset portfolio driven by the Black-Scholes model, and the performance analysis of communication systems. In this talk, we will discuss some of the interesting computational challenges that arise when a Gaussian copula is used to model the dependence, and what implications this has for modelling extreme events. We give some of the reasons why the Gaussian copula model has been blamed for financial mismanagement and how it may be reinstated in some limited number of applications. Technicalities will be kept to a minimum.
Date: Wednesday, 31 May 2017
6:00pm – 6:30pm: Refreshments
6:30pm – 7:30pm: Lecture
7:45pm onwards: Dinner (at a nearby restaurant)
Biography of Zdravko Botev
Zdravko is a DECRA fellow at UNSW. His research interests include: adaptive Monte Carlo methods for rare-event probability estimation, network reliability modeling, and kernel density estimation. He is the author of one of the most widely used Matlab software scripts for nonparametric density estimation and has written jointly with D. P. Kroese and T. Taimre the Handbook of Monte Carlo Methods, published by John Wiley & Sons in 2011.
|Time:||6:00 pm - 7:30 pm|
|Location:||UTS Maths Grid Room CB04.05.430,
Room 430, Level 5, Building 4, University of Technology,