EVENT DETAILS AND ABSTRACT

Title: Sequential Change Detection and Identification Problems for Stochastic Processes
Speaker: Professor Savas Dayanik
Speaker Info: Princeton University
Brief Description:
Special Note:
Abstract:

The characteristics of the stochastic models used to describe financial asset prices, hospital emergency room visits, computer network traffic are often expected to change at an unknown and unobservable time in the future, due to a shift in the economic cycle, a disease outbreak, or a cyber-attack. To hedge against the emerging new investment risks, protect the public health, prevent unauthorized access to computer network resources on a timely basis, one faces the problem of detecting those disorder times as quickly as possible while keeping the number of false alarms at acceptable low levels. We will show how these important problems can be formulated and solved in a Bayesian setting when the underlying stochastic model is a Levy process.