The TES methodology: Modeling empirical stationary time series

Benjamin Melamed, Jon R. Hill, David Goldsman

Research output: Chapter in Book/Report/Conference proceedingConference contribution

21 Scopus citations

Abstract

Autocorrelated processes occur naturally in many domains. Typical examples include autocorrelated (bursty) job arrivals to a manufacturing shc,p or telecommunications network. This paper presents a novel approach to input analysis of autocorrelated processes, called the TES ( Transform-Expand-Sample) modeling methodology. TES is a versatile class of stochastic processes which can simultaneously capture both the marginal clistribution and autocorrelation structure of a stationary (empirical) time series. In this paper we summarize the TES modeling methodology and briefly review a software environment, called TEStool, which' supports this methodology through an interactive graphical user interface (GUI). The GUI greatly facilitates the process of fitting a TES model to empirical time series, by providing immediate feedback to modeling actions. We conclude the paper with a number of examples which demonstrate the efficacy of the TES methodology and the TEStool GUI by fitting TES models to empirical datasets obtained from actual field measurements.

Original languageAmerican English
Title of host publicationProceedings of the 24th Conference on Winter Simulation, WSC 1992
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages135-144
Number of pages10
ISBN (Print)0780307984
DOIs
StatePublished - Dec 1 1992
Externally publishedYes
Event24th Conference on Winter Simulation, WSC 1992 - Arlington, United States
Duration: Dec 13 1992Dec 16 1992

Publication series

NameProceedings - Winter Simulation Conference

Other

Other24th Conference on Winter Simulation, WSC 1992
Country/TerritoryUnited States
CityArlington
Period12/13/9212/16/92

ASJC Scopus subject areas

  • Software
  • Modeling and Simulation
  • Computer Science Applications

Fingerprint

Dive into the research topics of 'The TES methodology: Modeling empirical stationary time series'. Together they form a unique fingerprint.

Cite this