A Monte Carlo method for estimating the correlation exponent

T. Mikosch*, Qiang Wang

*Corresponding author for this work

    Research output: Contribution to journalArticleAcademicpeer-review

    13 Citations (Scopus)

    Abstract

    We propose a Monte Carlo method for estimating the correlation exponent of a stationary ergodic sequence. The estimator can be considered as a bootstrap version of the classical Hill estimator. A simulation study shows that the method yields reasonable estimates.

    Original languageEnglish
    Pages (from-to)799-813
    Number of pages15
    JournalJournal of Statistical Physics
    Volume78
    Issue number3-4
    DOIs
    Publication statusPublished - Feb-1995

    Keywords

    • CORRELATION EXPONENT
    • CORRELATION DIMENSION
    • HILL ESTIMATOR
    • MONTE CARLO METHOD
    • BOOTSTRAP
    • TAIL INDEX
    • ATTRACTORS

    Cite this