Efficient fitting of long-tailed data sets into phase-type distributions

Authors:
Alma Riska;Vesselin Diev;Evgenia Smirni
Affiliations:
College of William and Mary, Williamsburg, VA;College of William and Mary, Williamsburg, VA;College of William and Mary, Williamsburg, VA
Venue:
ACM SIGMETRICS Performance Evaluation Review
Year:
2002

Citing 2
Cited 1

Fitting mixtures of exponentials to long-tail distributions to analyze network performance models

Performance Evaluation
MAMSolver: A Matrix Analytic Methods Tool

TOOLS '02 Proceedings of the 12th International Conference on Computer Performance Evaluation, Modelling Techniques and Tools

Cluster-based fitting of phase-type distributions to empirical data

Computers & Mathematics with Applications

Quantified Score

Hi-index	0.01

Visualization

Abstract

We propose a new technique for fitting long-tailed data sets into phase-type (PH) distributions. This technique fits data sets with non-monotone densities into a mixture of Erlang and hyperexponential distributions, and data sets with completely monotone densities into hyperexponential distributions. The method first partitions the data set in a divide and conquer fashion and then uses the Expectation-Maximization (EM) algorithm to fit the data of each partition into a PH distribution. The fitting results for each partition are combined to generate the final fitting for the entire data set. The new method is accurate, efficient, and allows one to apply existing analytic tools to analyze the behavior of queueing systems that operate under workloads that exhibit long-tail behavior, such as queues in Internet-related systems.