Search

This paper suggests a general class of estimators for unknown general population parameter of the study variable using auxiliary information in sample surveys. A large number of estimators of different parameters of the study variable based on auxiliary information can be viewed as a member of the suggested class of estimators. We have derived the bias and mean squared error of the suggested class of estimators under large sample approximation. Optimum conditions are obtained for which the proposed class of estimator has the minimum mean squared error. In particular, a class of estimators for the population mean of the study variable based on the coefficient variation of the auxiliary variable has been derived. The bias and mean squared error of the proposed class of estimators are obtained under large sample approximation. It has been shown that the proposed class of estimators is better than the usual unbiased estimator y ¯ and Das and Tripathi’s (Sankhya C 42(1–2):76–86, 1980) estimator in the context of bivariate normal and bivariate non-normal populations. In support of the present study, an empirical is given. [ABSTRACT FROM AUTHOR]

In many human surveys, information is in most cases not obtained from all the units in the survey even after some call backs. To cope with this problem, survey statisticians usually consider and adopt the non-respondents sub sampling schemes introduced by Hansen and Hurwitz (1946, Jour. Amer. Statist. Assoc., 41, 517–529). The topic which is of practical importance is the estimation of population mean at the current occasion in two- occasion successive sampling in presence of non-response. In this paper, an alternative to Satici and Kadilar (2011, Hacettepe Jour. Math. Statist. 40(1), 115–124) estimator for the population mean at the current occasion in successive sampling having two occasions is suggested under the missing data case. The minimum mean squared error and optimum fraction of a fresh sample of suggested estimator are obtained. The suggested estimator is compared with the estimator due to Singh and Priyanka (2007, Statistics in Transition. 8(2), 273–292) and Satici and Kadilar (2011, Hacettepe Jour. Math. Statist. 40(1), 115–124) and the efficient conditions for the suggested estimator are obtained. It has been shown theoretically that the proposed class of estimators is more efficient than the Singh and Priyanka (2007) and Satici and Kadilar (2011) estimators. The performance of the suggested class of estimators is examined over Singh and Priyanka (2007) and Satici and Kadilar (2011) estimators with the aid of an empirical study. [ABSTRACT FROM AUTHOR]

The crux of the paper is to develop a very general class of estimators for population mean Y ¯ of the study variable y using information on auxiliary variable x in simple random sampling. It has been shown that a large number of known and unknown estimators are members of the suggested class of estimators. Expressions of bias and mean squared errors of the proposed class of estimators have been obtained under large sample approximation. Asymptotic optimum estimator (AOE) in the proposed class of estimators has been obtained with its mean squared error formula. An empirical study is carried out to demonstrate the superiority of the members of the proposed class of estimators over existing estimators. [ABSTRACT FROM AUTHOR]