Leveraging the Rao-Blackwell theorem to improve ratio estimators in adaptive cluster sampling

Chang Tai Chao, Arthur L. Dryver, Tzu Ching Chiang

Research output: Contribution to journalArticlepeer-review

12 Citations (Scopus)


Rao-Blackwellization is used to improve the unbiased Hansen-Hurwitz and Horvitz-Thompson unbiased estimators in Adaptive Cluster Sampling by finding the conditional expected value of the original unbiased estimators given the sufficient or minimal sufficient statistic. In principle, the same idea can be used to find better ratio estimators, however, the calculation of taking all the possible combinations into account can be extremely tedious in practice. The simplified analytical forms of such ratio estimators are not currently available. For practical interest, several improved ratio estimators in Adaptive Cluster Sampling are proposed in this article. The proposed ratio estimators are not the real Rao-Blackwellized versions of the original ones but make use of the Rao-Blackwellized univariate estimators. How to calculate the proposed estimators is illustrated, and their performance are evaluated by both of the Bivariate Poisson clustered process and a real data. The simulation result indicates that the proposed improved ratio estimators are able to provide considerably advantageous estimation results over the original ones.

Original languageEnglish
Pages (from-to)543-568
Number of pages26
JournalEnvironmental and Ecological Statistics
Issue number3
Publication statusPublished - 2011 Sept

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Environmental Science(all)
  • Statistics, Probability and Uncertainty


Dive into the research topics of 'Leveraging the Rao-Blackwell theorem to improve ratio estimators in adaptive cluster sampling'. Together they form a unique fingerprint.

Cite this