Exclusive Content & Downloads from ASQ

Effectiveness of Bootstrap Bias Correction in the Context of Clock Offset Estimators

Summary: [This abstract is based on the authors' abstract.]Estimating and correcting the offset of clocks is important in data communication networks. The bias and mean squared error properties of most clock offset estimators depend on the distribution of network delays, yet no single distribution adequately characterizes these delays. The effectiveness of bootstrap bias correction of alternative estimators is examined under varying assumptions for network delays, including heavy-tailed distributions. Emphasis is given to Pareto distribution because of its applicability in Internet contexts. The bootstrap bias-corrected form of the Paxson estimator is shown to be more robust than other candidates considered. An unexpected result is that if network delays follow a Pareto distribution, bootstrap bias correction of both the exponential ordered-BLUE and the Pareto o-BLUE fails in the sense that the absolute bias increases.

Anyone with a subscription, including Site and Enterprise members, can access this article.

Other Ways to Access content:

Join ASQ

Join ASQ as a Full member. Enjoy all the ASQ member benefits including access to many online articles.

Subscribe to Technometrics

Access this and ALL OTHER Technometrics online articles. You'll also receive the print version by mail.

  • Topics:
  • Keywords: Bias,Bootstrap methods,Distribution,Mean square errors (MSE),Order statistics,Pareto charts,Robust design
  • Author: Jeske, Daniel R.; Chakravartty, Arunava
  • Journal: Technometrics