ESTIMATION OF FINITE POPULATION MEAN USING KNOWN COEFFICIENT OF VARIATION IN THE SIMULTANEOUS PRESENCE OF NON -RESPONSE AND MEASUREMENT ERRORS UNDER DOUBLE SAMPLING SCHEME 1 (original) (raw)

The present article considers the problem of finite population mean estimation, when the non-response and measurement errors are present simultaneously utilizing known information on coefficient of variation of study variable. We have developed an estimator of population mean which is improved and efficient, using Hansen and Hurwitz (1946) technique. Asymptotic expressions of the bias and variance of suggested estimator have been found correct up to approximation of degree one.The optimum value of characterizing scalar for which variance of proposed estimator is minimum has also been calculated. We have made a theoretical efficiency comparison of proposed estimator with usual Hansen Hurwitz estimator. To amply corroborate the theoretical findings, a simulation study has also been carried out using R software.