00000nam 2200000za 4500 9.840553 CaOODSP 20221107151855 cr ||||||||||| 170727s1988 onc |o f|0| 0 eng d CaOODSP eng eng fre n-cn--- CS11-617/88-27E-PDF MacMillan, J. H. A non-parametric empirical Bayes approach for estimating a process average in quality control [electronic resource] / by J. H. MacMillan and W. V. Mudryk. [Ottawa] : Statistics Canada, 1988. [14] p. Working paper ; 88-27 Digitized edition from print [produced by Statistics Canada]. "Working Paper No. BSMD-88-027E." "August 1988." Includes bibliographic references. "At Statistics Canada, acceptance sampling is used as a method of quality control for survey processing operations. The sampling plans which are used will ensure minimum inspection at a specific incoming error level. This error level is estimated by a quantity known as the process average. It is an unknown parameter which is usually estimated from current inspection results, but frequently the estimation is difficult because of small sample sizes. Greater accuracy in the estimate may be produced by using more data from previous samples to improve upon the current sample result. A non-parametric empirical Bayes estimator of the process average is presented. An approximate confidence interval is also constructed. Examples are provided"--Abstract. Abstract also in French. gccst Statistical analysis gccst Methodology Mudryk, W. V. Canada. Statistics Canada. Methodology Branch. Working paper (Statistics Canada. Methodology Branch) 88-27 (CaOODSP)9.834763 PDF 972 KB https://publications.gc.ca/collections/collection_2017/statcan/11-613/CS11-617-88-27-eng.pdf 11-617E no. 88-27