Browsing by Author "Hsing, Peter Shih-Shiang"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
- Orthogonal statistics involving the third and fourth sample moments for negative binomial distributionHsing, Peter Shih-Shiang (Virginia Tech, 1964-08-05)This thesis is an extension of the development of orthogonal statistics which can be used to investigate sampling properties of moment estimators. This work is particularized for estimators of parameters of the negative binomial distribution.
- Quality control operating procedures for multiple quality characteristics and weighting factorsHsing, Peter Shih-Shiang (Virginia Polytechnic Institute and State University, 1973)This research is devoted to the development of a procedure for manufacturing use that applies to cases involving a manufactured item having more than one quality characteristic and continuous measurements of these characteristics. There are two phases in this procedure. The first phase consists of testing the stability of the manufacturing process with respect to dispersion and central tendency. Once the stability of the process has been established, maintain the process within the established bounds of dispersion and central tendency becomes important. The second phase consists of monitoring dispersion and central tendency of the current manufacturing process. During the monitoring central tendency activity, the user is allowed to assign different weights to the sample mean value both above and below the standard value. For both dispersion and central tendency monitoring activities, a method is provided to identify quality characteristics or the interaction between quality characteristics as the cause of the trouble when the sampling indicated a lack of control. An example of the antidiarrheal tablet manufacturing process was presented to illustrate how to test the stability of the manufacturing process and how to monitor the current manufacturing process. Step-by-step computational procedures and Fortran computer programs were presented. The cases for dispersion and central tendency deviated from the standard values were simulated and the results were presented.