This thesis is about multiple hypothesis testing and its relation to the P-value. In Chapter 1, the methodologies of hypothesis testing among the three inference schools are reviewed. Jeffreys, Fisher, and Neyman advocated three different approaches for testing by using the posterior probabilities, P-value, and Type I error and Type II error probabilities respectively. In Berger's words ``Each was quite critical of the other approaches." Berger proposed a potential methodological unified conditional frequentist approach for testing. His idea is to follow Fisher in using the P-value to define the strength of evidence in data and to follow Fisher's method of conditioning on strength of evidence; then follow Neyman by computing Type I and Type II error probabilities conditioning on strength of evidence in the data, which equal the objective posterior probabilities of the hypothesis advocated by Jeffreys. Bickis proposed another estimate on calibrating the null and alternative components of the distribution by modeling the set of P-values as a sample from a mixed population composed of a uniform distribution for the null cases and an unknown distribution for the alternatives. For tackling multiplicity, exploiting the empirical distribution of P-values is applied. A variety of density estimators for calibrating posterior probabilities of the null hypothesis given P-values are implemented. Finally, a noninterpolatory and shape-preserving estimator based on B-splines as smoothing functions is proposed and implemented.

This thesis is about multiple hypothesis testing and its relation to the P-value. In Chapter 1, the methodologies of hypothesis testing among the three inference schools are reviewed. Jeffreys, Fisher, and Neyman advocated three different approaches for testing by using the posterior probabilities, P-value, and Type I error and Type II error probabilities respectively. In Berger's words ``Each was quite critical of the other approaches." Berger proposed a potential methodological unified conditional frequentist approach for testing. His idea is to follow Fisher in using the P-value to define the strength of evidence in data and to follow Fisher's method of conditioning on strength of evidence; then follow Neyman by computing Type I and Type II error probabilities conditioning on strength of evidence in the data, which equal the objective posterior probabilities of the hypothesis advocated by Jeffreys. Bickis proposed another estimate on calibrating the null and alternative components of the distribution by modeling the set of P-values as a sample from a mixed population composed of a uniform distribution for the null cases and an unknown distribution for the alternatives. For tackling multiplicity, exploiting the empirical distribution of P-values is applied. A variety of density estimators for calibrating posterior probabilities of the null hypothesis given P-values are implemented. Finally, a noninterpolatory and shape-preserving estimator based on B-splines as smoothing functions is proposed and implemented.

No description available.

Platinum-group elements (pGEs) are recovered from UG2 chromitite by milling and flotation. The mechanisms involved during beneficiation of this type of ore are still poorly understood, partly because of its complex nature. Image-analysis techniques were used to characterise the mineralogy ofUG2 chromitite from diverse geological environments, as well as the milling and flotation products derived from each of these ores. Postmagmatic alteration ofUG2 chromitite has a profound effect on the mineralogy, chemistry and recovery characteristics of the UG2 chromitite. Relatively unaltered UG2 chromitite consists predominantly of chromite and primary silicates, mostly bronzite and plagioclase with minor phlogopite, and small amounts of secondary silicates such as talc and chlorite. Trace quantities of base-metal sulphides, predominantly pentlandite, pyrrhotite and chalcopyrite ± pyrite, generally occur at chromite-silicate grain boundaries. PGEs are present both as discrete PGE minerals, and, to a lesser extent, sub-microscopically in other phases, mostly palladium and rhodium in pentlandite. The PGE mineral assemblage is characterised by sulphide minerals, mostly braggite, cooperite, nickeloan malanite and laurite, and is closely associated with the base¬metal sulphides. Recovery of PGE minerals is strongly dependent on the degree of liberation, with liberated PGE minerals and PGE minerals associated with liberated base-metal sulphides, the fastest-floating particles. PGE minerals report to flotation tailings predominantly as fine-grained inclusions in coarse silicate particles. In places, the footwall rocks have been replaced by iron-rich ultrabasic pegmatoid. As a result of interaction with Fe- and Ti-rich fluids, the chromite grains in the UG2 chromitite have been enlarged due to sintering, and the PGE mineral assemblage replaced by one consisting predominantly of laurite, Pt-Fe alloy and other non¬sulphide PGE minerals. The non-sulphide PGE mineral grains appear to be slower ¬floating than sulphide PGE minerals. Low temperature hydrothermal alteration appears to have caused relatively widespread alteration of the UG2 chromitite in some areas, resulting in corrosion and redistribution of sulphide minerals, as well as the replacement of primary magmatic silicates by secondary silicates such as pumpellyite, epidote, prehnite, albite, talc, chlorite and quartz. Ore from such areas are characterised by a base-metal sulphide assemblage consisting predominantly of millerite, chalcopyrite, and pyrite. Base¬metal sulphide and PGE minerals occur in fine-grained intergrowths with silicates, resulting in poor liberation. In the samples investigated, composite particles were often faster-floating than expected, at least partly due to the presence of naturally floatable talc. The effect of faulting on the mineralogy of the UG2 chromitite probably depends on distance from the fault zone, and possibly also timing of faulting, and can cause cataclasis of the ore. Where cataclasis occurred, broken mineral grains are cemented by secondary, hydrous silicates. Liberation of base-metal sulphides and PGE minerals are poor, and recoveries consequently very low. It was demonstrated that reasonable estimates of total PGE+Au recovery can be made from the mineralogical characteristics ofUG2 chromitite ore. Based on the mineralogy of ore from a specific area, provision can be made for appropriate adjustments to metallurgical flowsheets. / Thesis (PhD)--University of Pretoria, 2006. / Geology / Unrestricted

Abstract The more and more IC makers take the Research job in Silicon Valley, development in Taiwan and China, Design for Manufacturing in Taiwan and Marketing in China. There are plenty of complete Supplier Chains with organizing the companys of IC Resign, Foundry, Si-IP, Subcontracted Assembly & Test Service. The capacity offered is almost over 60% worldwide with very high technique and quality level. This is getting more dominantly important to provide turn-key service durin IDM shrinking trend. Meanwhile the great gap in the supply chain is expanding in IC testing design and manufacturing, which is the bottle neck of IC industry in either cost, technical or turn-around time. The research is subjected to get a workable business model of IC testing Design Service working with a right product Segmentation, Targeting, Positioning to compensate the gap. The research takes kinds of reference of Competitive Advantages, Competition Strategies, Co-petition Strategy, Competition, Strategy, and Disruptive Innovation to construct the research model. IC Testing Design is an innovative business model in Taiwan Industry. Conseguently the research collects opinions from high level managers of IC industry. It also does a case study by using the case of BEST-itech, which is the first supplier of IC Testing Service in Taiwan market. We could learn the SWOP to see how to provide the Testing Design Service in right positioning and strategy. The targeting customers are all in the Supply Chain of Virtual IDM, including IDM, Foundry, Design House, Design Service, Test & Assembly, ATE Vendor, FA House etc. During the resession of 2008, much more demand toward Virtual-IDM is firmed. The outsourcing demand of IC testing design is getting clear. ATE vendor¡¦s market scale growth is getting slow as well as the ASP. So the requirement and opportunity of IC Testing Design Service is happening. The research concludes the IC Testing Design is a workable and necessary sub-chain in IC-subcon supply. The Value Network is a good model to tell how to transfer the comptition to be the compensator or customers for all players in the supply chain. The research also concludes 4 segments necessary in the IC Testing Design business model. Those are 1. Tester Opt ionization of Utilization. 2. Device Interface Integrated with Testing Design. 3. Add-on Solution Implement in Testing Tech-Gap. 4. Test Floor Automation and SPC Testing. Hopefully this research could provide a good picture to start the innovative business model to help the IC industry.