Cover Image

Convergence of changepoint estimators

Let Xn1, ..., Xnn be an array of independent random vectors such that Xn1, ..., Xn[n[theta]] have distribution function F, and Xn[n[theta]]+1, ..., Xnn have distribution function G with F [not equal to] G. In this paper we propose an estimator [theta]n of the changepoint [theta] and show that n([theta]n-[theta]) = O(ln n) with probability one
Year of publication:
1992
Authors: Ferger, Dietmar ; Stute, Winfried
Published in:
Stochastic Processes and their Applications. - Elsevier, ISSN 0304-4149. - Vol. 42.1992, 2, p. 345-351
Publisher: Elsevier
Keywords: changepoint estimator exponential tail bound almost sure convergence
Type of publication: Article
Source:
Persistent link: https://www.econbiz.de/10008872869
Similar items by person