Modified Variance Estimators for Non Response Problems in Survey Sampling
M. A. Bhat *
Division of Agri -Stat F.O.H, Sher-e-Kashmir University of Agricultural Sciences and Technology of Jammu (SKUAST), Kashmir, India
T. A. Raja
Division of Agri -Stat F.O.H, Sher-e-Kashmir University of Agricultural Sciences and Technology of Jammu (SKUAST), Kashmir, India
S. Maqbool
Division of Agri -Stat F.O.H, Sher-e-Kashmir University of Agricultural Sciences and Technology of Jammu (SKUAST), Kashmir, India
S. A. Mir
Division of Agri -Stat F.O.H, Sher-e-Kashmir University of Agricultural Sciences and Technology of Jammu (SKUAST), Kashmir, India
Nageena Nazeer
Division of Agri -Stat F.O.H, Sher-e-Kashmir University of Agricultural Sciences and Technology of Jammu (SKUAST), Kashmir, India
*Author to whom correspondence should be addressed.
Abstract
In this present study, we have discussed the issue of presence and absence of non-response that we often face in survey estimation. We have suggested the estimators to estimate the finite population Variance in the absence and presence of non-response, using the linear combination of coefficient of skewness and quartiles as auxiliary information. The expression for mean square errors of suggested estimators has been derived up to the first order of approximation. The comparison of existing estimators with suggested estimators has been made through an numerical illustration to prove the efficiency of suggested estimators over existing estimators.
Keywords: Simple random sampling, bias, mean square error, skewness and quartiles efficiency