Effect of Heteroscedasticity on the Comparison of Variance Component Estimators Under Unbalanced Random Model | ||||
The Egyptian Statistical Journal | ||||
Article 7, Volume 37, Issue 1, June 1993, Page 110-121 | ||||
Document Type: Original Article | ||||
DOI: 10.21608/esju.1993.314835 | ||||
View on SCiNiTO | ||||
Author | ||||
B. Singh | ||||
Abstract | ||||
Expressions for moments and probability of negative values for truncated estimators of the group variance component are derived under unbalanced one-way random model with unequal error variances. These results are used to investigate the effects of heteroscedasticity on the comparison of group variance component estimators under balanced and unbalanced situations, numerically, or certain apriori parametric values. The computed results reveal that heteroscedasticity has no effect on the comparison of truncated estimators, however, the comparison of ANOVA with truncated estimators is affected combinedly by heteroscedasticity and unbalancedness. | ||||
Keywords | ||||
ANOVA Estimator; Heteroscedasticity; Truncated Estimators and Unbalancedness | ||||
Statistics Article View: 20 |
||||