This study proposes ratio estimate by adapting the estimate type of Ray and Singh [J. Ind. Stat. Assoc. 19 (1981) 147] to traditional and the other ratio-type estimate in simple random sampling in literature. Theoretically, mean square error (MSE) equations of all proposed ratio estimate are obtained and compared with each other. By these comparisons the conditions, which make each proposed estimate more efficient than the others, are found. The theoretical results are supported by a numerical illustration.
2003 Elsevier Inc. All rights reserved.