Search of good rotation patterns using exponential method of estimation in two-occasion successive sampling

This paper intends to put emphasis on the role of two auxiliary variables on both the occasions to improve the precision of estimates at current (second)occasion in two-occasion successive sampling. There are different situations when (i) the information on the single auxiliary variable z1 is readily available on both the occasions and positively correlated with the study variable y, (ii) the information on the single auxiliary variable z2 is readily available on both the occasions and negatively correlated with the study variable y, (iii) the information on two auxiliary variables (z1, z2) are available on both the occasions, with one auxiliary variable z1 is positively correlated and the other negatively correlated with the study variable y; these have been discussed through proposing three estimators (one in each situation) of the population mean Y‾$\bar Y$ on the current (second) occasion and analyzing their properties. The properties of the proposed estimators have been studied and compared with the sample mean estimator when there is no matching from the previous occasion and traditional successive sampling estimator, which is a linear combination of the means of the matched and unmatched portion of the sample at the current (second) occasion. Optimal replacement policy is discussed. Empirical studies are carried out to show the domination of the recommended estimators.