Covariance of regenerative mean and variance estimators for the Markov chains

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

Abstract

The covariance matrix that appears in the central limit theorem for the regenerative mean and standard deviation estimators has been expressed in a form that allow several conclusions to be reached. The off-diagonal term, representing the covariance between the point and standard deviation estimators, is independent of the return state chosen for blocking. The expression for the variance of the standard deviation estimator shows that the variance is increased by kurtosis in the partial-sum random variables. The variance does depend on the return state used for blocking, and an example shows that the state with the shortest mean return time can have the greatest variance.

Original languageEnglish (US)
Title of host publicationWinter Simul Conf Proc 1988
PublisherPubl by IEEE
Pages473-475
Number of pages3
ISBN (Print)0911801421, 9780911801422
DOIs
StatePublished - 1988
Externally publishedYes
EventWinter Simulation Conference Proceedings - 1988 - San Diego, CA, USA
Duration: Dec 12 1988Dec 14 1988

Publication series

NameWinter Simulation Conference Proceedings
ISSN (Print)0275-0708

Other

OtherWinter Simulation Conference Proceedings - 1988
CitySan Diego, CA, USA
Period12/12/8812/14/88

All Science Journal Classification (ASJC) codes

  • Software
  • Modeling and Simulation
  • Safety, Risk, Reliability and Quality
  • Chemical Health and Safety
  • Applied Mathematics

Fingerprint Dive into the research topics of 'Covariance of regenerative mean and variance estimators for the Markov chains'. Together they form a unique fingerprint.

Cite this