Title :
Multivariate arrival rate estimation using semidefinite programming
Author :
Papp, Dávid ; Alizadeh, Farid
Author_Institution :
Rutgers Center for Oper. Res., Piscataway, NJ, USA
Abstract :
An efficient method for the smooth estimation of the arrival rate of non-homogeneous, multi-dimensional Poisson processes from inexact arrivals is presented. The method provides a piecewise polynomial spline estimator. It is easily parallelized, and it exploits the sparsity of the neighborhood structure of the underlying spline space; as a result, it is very efficient and scalable. Numerical illustration is included.
Keywords :
optimisation; piecewise polynomial techniques; splines (mathematics); stochastic processes; Poisson processes; multivariate arrival rate estimation; piecewise polynomial spline estimator; semidefinite programming; Approximation methods; Mathematical model; Maximum likelihood estimation; Optimization; Polynomials; Spline;
Conference_Titel :
Simulation Conference (WSC), Proceedings of the 2011 Winter
Conference_Location :
Phoenix, AZ
Print_ISBN :
978-1-4577-2108-3
Electronic_ISBN :
0891-7736
DOI :
10.1109/WSC.2011.6147982
