Work

Optimal experiment design in a filtering context with application to sampled network data

Year: 2010

Type: article

Abstract: We examine the problem of optimal design in the context of filtering multiple random walks. Specifically, we define the steady state E-optimal design criterion and show that the underlying optimization problem leads to a second order cone program. The developed methodology is applied to tracking network flow volumes using sampled data, where the design variable corresponds to controlling the sampling rate. The optimal design is numerically compared to a myopic and a naive strategy. Finally, we relate our work to the general problem of steady state optimal design for state space models.

Source: The Annals of Applied Statistics

DOI: 10.1214/09-aoas283

Authors: Harsh Singhal, George Michailidis

Institution: University of Michigan

Language: English

Topic: Optimal Experimental Design Methods

Subfield: Management Science and Operations Research

Field: Decision Sciences

Domain: Social Sciences

Open Access status: bronze

Funder: National Science Foundation

Awards: Extremes: Short and Long-Range Dependence; Modeling and Inference with Applications to Computer Networks and Risk Analysis, Design, Analysis, and Monitoring of Network Characteristics Using Tomography Techniques