P. Diko, M. Usábel
A general Lévy diffusion process – the solution to the Lévy stochastic differential equation - is treated in a stochastic control framework. All the characteristics of the process are subject to the control variable. The objective function is maximized in a fixed finite time horizon. The value function of the stochastic control problem is approximated through randomization of the horizon and assuming piecewise constant family of controls. It is shown that the procedure is convergent and unveils not only the value function but also the corresponding optimal control. The relevance of the procedure is demonstrated in applications to the portfolio selection problem and optimal execution time of an American put option.
Palabras clave: stochastic control, Lévy diffusion, numerical methods, randomization
Programado
VB7 Probabilidad, convergencias y teoremas límite
20 de abril de 2012 10:30
Sala Roma II
