Date of Submission

2-28-1974

Date of Award

2-28-1975

Institute Name (Publisher)

Indian Statistical Institute

Document Type

Doctoral Thesis

Degree Name

Doctor of Philosophy

Subject Name

Mathematics

Department

Research and Training School (RTS)

Supervisor

Hanurav, T. V. (RTS-Kolkata; ISI)

Abstract (Summary of the Work)

Probability enmpling methods ha ve extenaive a pplication in la rge- eca le agricultural and socio economic surve ys, Sampling is ale used in one or more stages of a census and sometimes as a mubstitute for a complete Census. Over the pa st four deca de s, confidence in the use of san pling and a ppreciation for the technique have stee dily grown thanks to the pi onee ring efforts of Neyman, Hansen, Hurwit z, Maha lanobis, Sukha tme, Cochran, Yate s and others.2. It is well known that the principa i problem of the theory of sampling from fin ite popu latione consists in deviring a suitable sample se le cti on procedure for aecuring a representative sample from a given population and de ve loping an appropriate estimator tor the population pa rameter of inte re at euch as the meen or total of a characterietic or ratio of totals of two cha racterietics, with a view to maximising the preciaion of the estimator within the avéila ble reBources or minimising the cost Icr a chieving a given le vel of pracialon.3. Towa rde the solution of the prinelpal problem of sampling from finite populations, the firat significant attempt appeara to be thit of Neyman (193 4) who laid down the principles goverming atratifie d random sampling. De ve lopment of cluster sampling may be tra ce d to the work of Hansen and Hurwitz (1942). Mu itista ge sampling technique and their applications may be found in the contributions by Hangen and Hurwitz (1943), Sukhatme (1947) and othe ra. Systematic sampling was Arst inveatigated hy Madow and Madow (1944). Use of auxilia ry infor- mation in sample sele ction was eugge sted by Han sen and Hurwitz (1943).Coch ran (1942) utiliend a uxilia ry information in ratio and regression methods of e atimation. The important concept ofcost function was intraduced by Maha lanobis (1940). A number of revie wa have a ppeare d on the significant de ve lopments in theoretical and a pplied re sta rch on sampling from finite population s. Special mention may be ma de of the reviews by Yate a (1946), Cochran (1947) Ste phen (19 48), Seng (1951). Sukhatme (1959, 1966). Da leniu a (1962) and Murthy (1962). 4. Heuristic conside rations seem to ha ve played a major role in the earlier deve lopmenta in devising repre sentative methode of sample se le ction and, in gene ral, efforts ha ve been limite d to construct some simple unbia se d or bia sed estimators and atudy their relative e fficiency. Though a number of an mple de signa and estimation methods have become a vailable, the problem of choosing an optimum st rategy ( sampling deoign together with an estimator being called a strategy) has not become ea sje r Recogni sing the need to formulate the theory of sampling from finite popu lations in a more systematic manner, Horvitz and Thompaon (1952) ha ve initis te da fruitfu! Iinn of re search and de fine d th ree cla sae of

Comments

ProQuest Collection ID: http://gateway.proquest.com/openurl?url_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:dissertation&res_dat=xri:pqm&rft_dat=xri:pqdiss:28842805

Control Number

ISILib-TH85

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

DOI

http://dspace.isical.ac.in:8080/jspui/handle/10263/2146

Download

Included in

Mathematics Commons

Share

COinS