The problem of estimating the coefficient of variation (?) is considered when it is a priori suspected that ?=?,. Estimators based on the pre-test principle are proposed, and their large-sample properties are studied and compared. The relative dominance of the estimators is presented. Interestingly, the proposed estimators perform better than the classical estimator over a wider range. More importantly, the size for the preliminary test of the estimators is reasonable. An example is also given.
No takes yet. Share an insight, caveat, or question.
S. Ejaz Ahmed (1994) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: