Conditioning as disintegration

Conditional probability distributions seem to have a bad reputation when it comes to rigorous treatment of conditioning. Technical arguments are published as manipulations of Radon–Nikodym derivatives, although we all secretly perform heuristic calculations using elementary definitions of conditional probabilities. In print, measurability and averaging properties substitute for intuitive ideas about random variables behaving like constants given particular conditioning information.
One way to engage in rigorous, guilt-free manipulation of conditional distributions is to treat them as disintegrating measures—families of probability measures concentrating on the level sets of a conditioning statistic. In this paper we present a little theory and a range of examples—from EM algorithms and the Neyman factorization, through Bayes theory and marginalization paradoxes—to suggest that disintegrations have both intuitive appeal and the rigor needed for many problems in mathematical statistics.     

The impact of skewness in the hedging decision

The impact of skewness in the hedger's objective function is tested using a model of hedging derived from a third‐order Taylor Series approximation of expected utility. To determine the effect of price skewness upon hedging and speculation, analytical results are derived using an example of cotton storage. Findings suggest that when forward risk premiums and price skewness in the spot asset have opposite signs, speculation increases relative to the mean‐variance model. When the signs are identical, speculation will decrease, contradicting findings of mean‐variance models. © 2006 Wiley Periodicals, Inc. Jrl Fut Mark 26:503–520, 2006     

我的九条命

在我生命中各个阶段里，我与学术界不知不觉地频频有缘。启蒙时期，我科到名师的熏导业已浸淫于国学；经史子籍，无所不读，撰文作诗，直追唐宋，俨然是一个“小老学究”。当我入仕时，一秉传统士大夫的志趋，一心为国为民，在我自我流放期间，发现我唯一可谋生的本领是学术，我既没有继承万贯家财，也没有在当权时放出许多人情，作为经商的捷径，实际上，我曾试探经商之路，但是在没有有力的后台引见下，我意识到，要加入一家企业，条件是能够为其带来多少商机，我对这种用才标准并不欣赏，在此情形下，要做一个小企业家，其生活方式也难以适应，特别是对我的妻女是一种冒险。     

Product Development Strategy and Organizational Learning: A Tale of Two PC Makers

The personal computer (PC) marketplace in the US presents a dizzying array of component suppliers and products. No single firm dominates the industry with a complete package of hardware and software components. Although one company's operating systems and general-purpose applications are installed on most PCs in the US, the other system components—processors, memory, storage devices, display adapters, monitors, specialized applications, and so on—come from any number of sources. David T. Methe, Ryoko Toyama, and Junichiro Miyabe point out that the PC industry in Japan also exhibits this decentralized nature. However, they also note that despite the decentralized network structure of the Japanese PC industry, one company—NEC—was able to achieve a dominant market share. To provide insight into the key issues involved in the management of complex technology, they contrast NEC's strategic approach to product development and organizational learning with the approaches taken by Fujitsu—the firm that placed a distant second in this market. Despite matching NEC in terms of technological capabilities, financial resources, and managerial talent, Fujitsu never managed to threaten NEC's dominance of the PC market in Japan. Fujitsu continually emphasized technological leadership, even at the expense of protecting its installed base. Poor coordination of resources and product development efforts resulted in incompatibilities among Fujitsu's various products, and the company failed to foster close relationships with suppliers of such key technologies as software and peripherals. NEC's PCs did not enjoy the advantages of first-to-market status or technological leadership. Instead, NEC achieved market dominance by finding the combination of product technologies that met the needs of the greatest number of consumers. Throughout almost 20 years of competition in the PC industry, NEC successfully maintained consistency and backward compatibility across its product lines. NEC also recognized the importance of third-party software developers, and carefully cultivated relations with these firms as a source of competitive advantage. In other words, NEC struck the right balance between three key factors: technological innovation, motivation of third-party developers of software and peripherals, and service to its installed base of customers.     

The case of the unpopular pay plan

Three years after launching the team-based Quality For All program, Top Chemical Company CEO Sam Verde was searching for a team-based compensation system that would reflect his company's new philosophy. With a committee gathered to discuss the issue, Verde confronts the fact that changing pay plans is an issue few people can agree on. "Very simply," explains vice president for compensation Gilbert Porterfield, "the plan is designed to give employees working on teams real incentives for constant improvement and overall excellence. The variable aspect of the system pays employees for the performance of their group." This doesn't sit well with the others. "It's going to punish teams like mine for the failings of others instead of rewarding us for the work we do and have already done," says packaging team representative Ruth Gibson. Another committee member feels that team-based anything is a "motivational happy land that doesn't square with how people really work." While Verde likes the proposed pay plan, he has doubts over whether his employees will accept the risk. Upper management has no problem basing 60% of its pay on TopChem's performance. But getting line employees to risk part of their salaries--even as little as 4%--on the ups and downs of the chemical industry may be more trouble than it's worth. Four experts on compensation reveal where Top Chemical went wrong in its plan and how Sam Verde might bring about change successfully.     

The savings potential of 24-hour coverage

In summary, there are costs to maintaining separate systems to cover both work- and non-work-related injuries and illnesses; there are also significant costs associated with achieving coordination--if not integration--of the two plans. Overall, the financial data do not indicate that the overlap between workers' compensation and health benefits is of such magnitude as to justify integration regardless of cost; however, the data do suggest that judicious exploitation of opportunities to coordinate the two programs, especially in regard to managing health care providers, may generate significant savings.