On the comparison of some non-negative estimators of variance components for two models
- 1 January 1984
- journal article
- research article
- Published by Taylor & Francis in Communications in Statistics - Simulation and Computation
- Vol. 13 (5) , 619-633
- https://doi.org/10.1080/03610918408812401
Abstract
In this paper, the estimators , ASR (P.S.R.S. Rao and Chaubey (1978)), CMINQUE (Chaubey (1983)) and some other modifications of MINQUE similar to that of J.N.K. Rao and Subrahmaniam (1971) are considered for two models (i) The common mean model with heteroscedastic variances and (ii) One way random effects model. These estimators are compared respect to their biases and efficiencies through a Monte Carlo Study. For model (i) weighted means based on different estimators of variances are also compared.Keywords
This publication has 11 references indexed in Scilit:
- Nonnegative Minimum Biased Invariant Estimation in Variance Component ModelsThe Annals of Statistics, 1981
- Estimators for the One-Way Random Effects Model with Unequal Error VariancesJournal of the American Statistical Association, 1981
- Mean Square Error Comparison for MINQUE, ANOVA and Two Alternative Estimators Under the Unbalanced One‐Way Random ModelBiometrical Journal, 1981
- Three modifications of the principle of the minqueCommunications in Statistics - Theory and Methods, 1978
- Comparison of Estimators of Heteroscedastic Variances in Linear ModelsJournal of the American Statistical Association, 1975
- On the Estimation of Heteroscedastic VariancesPublished by JSTOR ,1973
- Estimation of Variance and Covariance Components in Linear ModelsJournal of the American Statistical Association, 1972
- Combining Independent Estimators and Estimation in Linear Regression with Unequal VariancesPublished by JSTOR ,1971
- Estimation of Heteroscedastic Variances in Linear ModelsJournal of the American Statistical Association, 1970
- A Sampling Investigation of the Efficiency of Weighting Inversely as the Estimated VarianceBiometrics, 1953