Development of novel separate ratio estimator in stratified random sampling with applications on real data

In this study, we have developed an improved separate ratio estimator using the highly related variable for the estimation of population average of study variable under stratified sampling. Due to the complex nature of the equation obtained, we used Taylor's series approximation up to first degree for obtaining bias and mean square error. Empirical studies on the real data sets were utilised to check its relative efficiency in comparison with the estimators developed by several prominent authors. From both the empirical studies our proposed estimator shows gain in relative precision over the others.