In survey sampling we are often concerned with the estimation of population parameters with the use of auxiliary information, which may be available (or made available by diverting a part of the resources) in one form or the other. If used properly, this information may provide better estimates than those where such information is not used. In this context improved classes of estimators of population mean with the use of auxiliary information, improved ratio and product type of estimators have been developed for estimation of population mean of study variable .The families of estimators of mean have been proposed for various situations of auxiliary characters such as when variance is known and population mean unknown, when variance is unknown and population mean known and neither population mean nor variance is known. Further various classes of improved ratio type estimators and product type estimators have been developed. The large sample properties of the proposed estimators of mean, ratio and product type estimators have been compared with conventional and other estimators through empirical study for different sample sizes.