CFTC logo

Past seminars Seminários já decorridos

Predictability in Stochastic Processes

By: Frank Raischel
From: Universidade de Lisboa
At: Instituto de Investigação Interdisciplinar, Anfiteatro
[2014-10-30]

Stochastic Processes are ubiquitous in complex and nonequilibrium systems. Here, we will consider what the "stochastic" property implies for the predictability of such systems. A method is presented to increase predictability in certain stochastic processes that are described by multidimensional Langevin equations. Application of this method to time series of air pollutant concentrations is given, and recent results will illustrate the usefulness of stochastic processes for understanding wind energy fluctuations and financial ratings.