Date of Submission


Date of Award


Institute Name (Publisher)

Indian Statistical Institute

Document Type

Doctoral Thesis

Degree Name

Doctor of Philosophy

Subject Name



Research and Training School (RTS)


Rao, C. Radhakrishna (RTS-Kolkata; ISI)

Abstract (Summary of the Work)

This thesis consists of nine chapters. In the first chapter we give the basic concepts and definitions and also a broad review of the literature related to the problems considered in this thesis. The second and third chapters are devoted to a detailed modified comparison of sampling with and without replacement, for the case of equal and unequal probability sampling respectively. In the fourth and fifth chapters we discuss the criteria of hyper-admissibility and linear suf2iciency for the choice of an optimum estimator for a given sampling design. The succeeding three chapters have as their main objective the central problem of choice of an optimum sampling strategy subject to some cost restriction the last chapter we give some results concerning. Horvitz-Thompson estimator and its variance estimation. Since each chapter has its own summary at the beginning, we shall not give a detailed chapter-wine summary here. Instead We shall briefly describe the course of development of the theory of survey sampling to present date.Though the concept of sampling has been in vogue from time immemorial, it is only during the thirties and forties that a more systematic development of the theory of sample surveys has taken place with the advent of sampling without replacement, probability 'sampling and the theory of stratification. But the large number of techniques sometimes ingenious - developed and practised during that period were mostly based on empirical and intuitive considerations. It is only of late that attention has been paid to the purely theoretical aspects of the subject.A clear formulation of the central problem of the theory of sampling from finite populations is due to Godambe [14]. In that paper he proposed a unified theoretical set-up in which most of the problems of sampling from finite populations could be discussed. Since then tremendous progress has been made towards developing a unified theory of sampling. Whether or not the present status of the theory is satisfactory is a controversial issue and we propose not to enter into a discussion about it here.


ProQuest Collection ID:

Control Number


Creative Commons License

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


Included in

Mathematics Commons