Estimating the variances of ratio estimates in complex sample surveys with two primary sampling units per stratum—a comparison of balanced replication and jackknife techniques
- 1 January 1979
- journal article
- research article
- Published by Taylor & Francis in Journal of Statistical Computation and Simulation
- Vol. 8 (3-4) , 191-205
- https://doi.org/10.1080/00949657908810266
Abstract
The balanced half-sample and jackknife variance estimation techniques are used to estimate the variance of the combined ratio estimate. An empirical sampling study is conducted using computer-generated populations to investigate the variance, bias and mean square error of these variance estimators and results are compared to theoretical results derived elsewhere for the linear case. Results indicate that either the balanced half-sample or jackknife method may be used effectively for estimating the variance of the combined ratio estimate.Keywords
This publication has 4 references indexed in Scilit:
- Properties of the balanced half-sample and jackknife variance estimation techniques in the linear caseCommunications in Statistics - Theory and Methods, 1977
- Balanced Repeated Replications for Standard ErrorsJournal of the American Statistical Association, 1970
- NOTES ON BIAS IN ESTIMATIONBiometrika, 1956
- THE DESIGN OF OPTIMUM MULTIFACTORIAL EXPERIMENTSBiometrika, 1946