Loading…

Two stage successive sampling with partial replacement of units

Multistage sampling design are practically inevitable in any large scale surveys. In the available literature the attention has, by and large been given to equal sized first stage units (fsu) with same replacement fraction within the fsu because of algebric simplicity. However, in the practical situ...

Full description

Saved in:
Bibliographic Details
Published in:Journal of the Indian Society of Agricultural Statistics 2003-08, Vol.56 (2)
Main Authors: Gola, R.C, Rai, A. (Indian Agricultural Statistics Research Institute, New Delhi (India))
Format: Article
Language:English
Subjects:
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Multistage sampling design are practically inevitable in any large scale surveys. In the available literature the attention has, by and large been given to equal sized first stage units (fsu) with same replacement fraction within the fsu because of algebric simplicity. However, in the practical situations often met with, the fsu are generally of unequal size. Therefore, in this article partial matching among fsu as well as ssu has been considered for estimation of mean on the second occasion under two stage sampling designs. Three estimators of mean have been proposed for this purpose (i) General linear estimator (ii) Linear composite estimator and (iii) Ratio type composite estimator. Further, the above three estimators were compared empirically with the help of real survey data. It has been observed that the gain in efficiency for general linear estimator was of the order of 200% and for the ratio type composite estimator, the gain was only marginal. Thus it may be concluded that the sustantial gains in the efficiency of the estimators may be achieved by making use of thetechnique of successive sampling coupled with an appropriate choice of the type of estimator to be used.
ISSN:0019-6363