Abstract
In statistical inference one usual assumption is, that data relates to a set of independent identically distributed random variables. From the viewpoint of sampling theory this assumption is only satisfied, if we draw a simple random sample with replacement or the population size is infinite. Then it is not necessary to consider a finite population correction when calculating the variance of a given estimator. To examine the effect of simple random sampling without replacement on the above assumption, the exact variances are calculated in the cases of mean value and variance estimation. This may give us information whether finite population correction is neglible or not.
Similar content being viewed by others
Literatur
ARBEITSGEMEINSCHAFT MEDIA-ANALYSE E. V. (1980), Media-Analyse 1980—Erläuterungen und Dokumentation, Media-Micro-Census GmbH, Auftragsforschung und Verlag, Frankfurt a. M.
RAJ, D. (1968), Sampling theory, McGraw-Hill, New York.
RANDLES, R. H., WOLFE, D. A. (1979), Introduction to the theory of nonparametric statistics, Wiley, New York.
ROBSON, D. S. (1957), Application of multivariate polykays to the theory of unbiased ratio-type estimation, JASA52, p. 511–522.
SUKHATME, P. V. (1939), On bipartional functions, Phil. Trans. Roy. Soc. Lond. A237, p. 375–409.
SUKHATME, P. V. (1943), Moments and product moment-statistics for samples of the finite and infinite populations, Sankhyā6, p. 363–382.
SKHATME P. V., SKHATME, B. V. (1970), Sampling theory of surveys with applications, 2nd ed., Iowa State University Press, Ames, Iowa.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Kreienbrock, L. Zur Auswirkung von Endlichkeitskorrekturen bei der Analyse einfacher Zufallsstichproben. Statistische Hefte 27, 23–35 (1986). https://doi.org/10.1007/BF02932553
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF02932553