A Benchmark for PH Estimation Algorithms: Results for Acyclic-PH

Andrea Bobbio, M. Telek

Research output: Contribution to journalArticle

56 Citations (Scopus)

Abstract

Distribution functions that can be expressed as exponential polynomials have useful computational properties in applied stochastic modeling and have gained widespread acceptance in recent years. Nevertheless, the implementation of efficient numerical procedures for estimating the distribution parameters remains an open problem that limits the use of this class of distributions in applications. The difficulty of the fitting problem is largely related to the non-linearity of the model and to the number of the parameters to be estimated. Many attempts have been presented in the literature. However, the lack of accepted and standardized test examples makes it difficult to establish a comparative merit among the various approaches. This paper proposes a benchmark based on the workshop on Fitting phase type distributions, organized by S. Asmussen in February 1991. It also presents the results obtained by applying the Maximum Likelihood (ML) estimation procedure to the canonical representation of Acyclic Phase Type (APH) distributions.

Original languageEnglish
Pages (from-to)661-677
Number of pages17
JournalCommunications in Statistics. Part C: Stochastic Models
Volume10
Issue number3
DOIs
Publication statusPublished - 1994

Fingerprint

Phase-type Distribution
Maximum likelihood estimation
Estimation Algorithms
Distribution functions
Polynomials
Benchmark
Exponential Polynomial
Canonical Representation
Stochastic Modeling
Numerical Procedure
Maximum Likelihood Estimation
Open Problems
Distribution Function
Nonlinearity
Model
Class

Keywords

  • Acyclic PH distributions
  • Maximum likelihood estimation
  • PH distributions

ASJC Scopus subject areas

  • Applied Mathematics
  • Statistics and Probability
  • Modelling and Simulation

Cite this

A Benchmark for PH Estimation Algorithms : Results for Acyclic-PH. / Bobbio, Andrea; Telek, M.

In: Communications in Statistics. Part C: Stochastic Models, Vol. 10, No. 3, 1994, p. 661-677.

Research output: Contribution to journalArticle

@article{8dfa8f0a7dbc421e94402da29e1f84a2,
title = "A Benchmark for PH Estimation Algorithms: Results for Acyclic-PH",
abstract = "Distribution functions that can be expressed as exponential polynomials have useful computational properties in applied stochastic modeling and have gained widespread acceptance in recent years. Nevertheless, the implementation of efficient numerical procedures for estimating the distribution parameters remains an open problem that limits the use of this class of distributions in applications. The difficulty of the fitting problem is largely related to the non-linearity of the model and to the number of the parameters to be estimated. Many attempts have been presented in the literature. However, the lack of accepted and standardized test examples makes it difficult to establish a comparative merit among the various approaches. This paper proposes a benchmark based on the workshop on Fitting phase type distributions, organized by S. Asmussen in February 1991. It also presents the results obtained by applying the Maximum Likelihood (ML) estimation procedure to the canonical representation of Acyclic Phase Type (APH) distributions.",
keywords = "Acyclic PH distributions, Maximum likelihood estimation, PH distributions",
author = "Andrea Bobbio and M. Telek",
year = "1994",
doi = "10.1080/15326349408807315",
language = "English",
volume = "10",
pages = "661--677",
journal = "Stochastic Models",
issn = "1532-6349",
publisher = "Taylor and Francis Ltd.",
number = "3",

}

TY - JOUR

T1 - A Benchmark for PH Estimation Algorithms

T2 - Results for Acyclic-PH

AU - Bobbio, Andrea

AU - Telek, M.

PY - 1994

Y1 - 1994

N2 - Distribution functions that can be expressed as exponential polynomials have useful computational properties in applied stochastic modeling and have gained widespread acceptance in recent years. Nevertheless, the implementation of efficient numerical procedures for estimating the distribution parameters remains an open problem that limits the use of this class of distributions in applications. The difficulty of the fitting problem is largely related to the non-linearity of the model and to the number of the parameters to be estimated. Many attempts have been presented in the literature. However, the lack of accepted and standardized test examples makes it difficult to establish a comparative merit among the various approaches. This paper proposes a benchmark based on the workshop on Fitting phase type distributions, organized by S. Asmussen in February 1991. It also presents the results obtained by applying the Maximum Likelihood (ML) estimation procedure to the canonical representation of Acyclic Phase Type (APH) distributions.

AB - Distribution functions that can be expressed as exponential polynomials have useful computational properties in applied stochastic modeling and have gained widespread acceptance in recent years. Nevertheless, the implementation of efficient numerical procedures for estimating the distribution parameters remains an open problem that limits the use of this class of distributions in applications. The difficulty of the fitting problem is largely related to the non-linearity of the model and to the number of the parameters to be estimated. Many attempts have been presented in the literature. However, the lack of accepted and standardized test examples makes it difficult to establish a comparative merit among the various approaches. This paper proposes a benchmark based on the workshop on Fitting phase type distributions, organized by S. Asmussen in February 1991. It also presents the results obtained by applying the Maximum Likelihood (ML) estimation procedure to the canonical representation of Acyclic Phase Type (APH) distributions.

KW - Acyclic PH distributions

KW - Maximum likelihood estimation

KW - PH distributions

UR - http://www.scopus.com/inward/record.url?scp=0003141975&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0003141975&partnerID=8YFLogxK

U2 - 10.1080/15326349408807315

DO - 10.1080/15326349408807315

M3 - Article

AN - SCOPUS:0003141975

VL - 10

SP - 661

EP - 677

JO - Stochastic Models

JF - Stochastic Models

SN - 1532-6349

IS - 3

ER -