Two statistical coverage problems in estimating the variance of a population
Author(s) -
Wiebe R. Pestman
Publication year - 2013
Publication title -
elemente der mathematik
Language(s) - English
Resource type - Journals
eISSN - 1420-8962
pISSN - 0013-6018
DOI - 10.4171/em/234
Subject(s) - variance (accounting) , statistics , mathematics , population , econometrics , demography , economics , accounting , sociology
Suppose a researcher draws a sample X1, X2, . . . , Xm from some population, and computes the corresponding variance in it. This in order to estimate the variance of the population from which the sample was drawn. Assume that the population in question has a Gaußian probability distribution. A second researcher draws, independently, a sample Y1,Y2, . . . ,Yn from the same population. He computes not only the corresponding variance in the sample, but also surrounds it by margins such as to get a 95% confidence interval for the population variance. Then what is the probability that this 95% confidence interval, generated by Y1,Y2, . . . ,Yn , will cover the sample variance of X1, X2, . . . , Xm? Below this probability will be denoted by Pm n . As a second coverage problem, what is the probability that the 95% confidence interval generated by the X1, X2, . . . , Xm and that by the Y1,Y2, . . . ,Yn are disjoint? Below this probability will be denoted by Qn . The aim
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom