Abstract
Let ξ1, ξ2, ξ3, ... be independent identically distributed random variables each with normal distribution with mean μ and variance σ2. Tests for the process mean μ are well-known elements of statistical analysis: the Gauß test under known process variance σ2, Student’st-test under unknown process variance σ2. Let the process be partitioned in lots (ξ1, ..., ξ N ), (ξ N+1, ..., ξ2N ), ... of sizeN. Consider (ξ1, ..., ξ N ) as a stochastic representative of this lot sequence and let the lot be characterized by the lot mean\(\frac{1}{N}\sum\limits_{i = 1}^N {\xi _i } \). The lot mean can be considered as a parameter of the joint conditional distribution function of the lot variables under\(\frac{1}{N}\sum\limits_{i = 1}^N {\xi _i } = z\). The present paper investigates the analogies of the Gauß test and Student’st-test for the lot situation, i.e. tests of significance for the lot meanz under known and unknown process variance σ2. This approach is of special interest for the statistical control of product quality in situations where the quality of a lot of items 1, 2, ...,N with quality characteristics ξ1, ξ2, ..., ξ N is identified with the lot average\(\frac{1}{N}\sum\limits_{i = 1}^N {\xi _i } = z\).
Similar content being viewed by others
References
Göb R (1996) An elementary model for statistical lot inspection and its application to sampling by variables. Metrika 44:135–163
Lehmann EL (1959) Testing statistical hypotheses. John Wiley, New York
Rényi A (1970) Probability Theory. North-Holland Publishing Company, Amsterdam/London
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Göb, R. Tests of significance for the mean of a finite lot. Metrika 44, 223–238 (1996). https://doi.org/10.1007/BF02614068
Received:
Issue Date:
DOI: https://doi.org/10.1007/BF02614068