Cluster‐Specific Variable Selection for Product Partition Models

We propose a random partition model that implements prediction with many candidate covariates and interactions. The model is based on a modified product partition model that includes a regression on covariates by favouring homogeneous clusters in terms of these covariates. Additionally, the model allows for a cluster‐specific choice of the covariates that are included in this evaluation of homogeneity. The variable selection is implemented by introducing a set of cluster‐specific latent indicators that include or exclude covariates. The proposed model is motivated by an application to predicting mortality in an intensive care unit in Lisboa, Portugal.     

Runtime and memory consumption analyses for machine learning R programs

R is a multi-paradigm language with a dynamic type system, different object systems and functional characteristics. These characteristics support the development of statistical algorithms at a high level of abstraction. Although R is commonly used in the statistics domain a big disadvantage are its runtime problems when handling computation-intensive algorithms. Especially in the domain of machine learning the execution of pure R programs is often unacceptably slow. Our long-term goal is to resolve these issues and in this contribution we used the traceR tool to analyse the bottlenecks arising in this domain. Here we measured the runtime and overall memory consumption on a well-defined set of classical machine learning applications and gained detailed insights into the performance issues of these programs.     

Sample size planning for phase II trials based on success probabilities for phase III

In recent years, high failure rates in phase III trials were observed. One of the main reasons is overoptimistic assumptions for the planning of phase III resulting from limited phase II information and/or unawareness of realistic success probabilities. We present an approach for planning a phase II trial in a time‐to‐event setting that considers the whole phase II/III clinical development programme. We derive stopping boundaries after phase II that minimise the number of events under side conditions for the conditional probabilities of correct go/no‐go decision after phase II as well as the conditional success probabilities for phase III. In addition, we give general recommendations for the choice of phase II sample size. Our simulations show that unconditional probabilities of go/no‐go decision as well as the unconditional success probabilities for phase III are influenced by the number of events observed in phase II. However, choosing more than 150 events in phase II seems not necessary as the impact on these probabilities then becomes quite small. We recommend considering aspects like the number of compounds in phase II and the resources available when determining the sample size. The lower the number of compounds and the lower the resources are for phase III, the higher the investment for phase II should be. Copyright © 2015 John Wiley & Sons, Ltd.     

Comparison of conflict of interest policies and reported practices in academic medical centers in the United States

The authors reviewed the conflict of interest policies of 9 academic medical centers in the United States and interviewed members of the Institutional Review Boards (IRBs) and Conflict of Interest Committees (COICs) at those institutions. They found that many institutions used processes for reporting and managing conflicts of interest that were more decentralized than the processes described in their policies. Also, most institutions had no clear and comprehensive policy to guide investigators regarding disclosure of conflicts of interest to potential research participants. Considerable differences in understanding of conflict of interest policies were observed between IRB and COIC officials.     

Market Orientation,Ownership Type,and E‐Business Assimilation: Evidence from Chinese Firms

While more and more firms have implemented e‐business in business operations, a better understanding of the factors that successfully drive the assimilation of e‐business will provide insights for firm executives and practitioners to develop effective strategies for e‐business. Different from previous studies that focus on individual‐level factors related to business executives and top management teams, this study examines how firm‐level strategic and cultural factors shape e‐business assimilation. Based on the strategy and marketing literature on market orientation and firm ownership, we developed a research model to describe how a firm's market orientation impacts e‐business assimilation. The model also describes the moderating effect of firm ownership type on the relationship between market orientation and e‐business assimilation. Based on data from 301 Chinese international trade firms, we found that two dimensions of market orientation (i.e., customer orientation, competitor orientation) had significant effects on e‐business assimilation. However, the third dimension, interfunctional coordination, was only partially significant. In addition, ownership type was a significant moderator of the effects of customer orientation and competitor orientation on e‐business assimilation, although ownership type was not a moderator of interfunctional coordination. Being one of the first studies of the impact of market orientation and firm ownership type on e‐business assimilation, we conclude with a discussion of the implications for future research and practice.     

SENS and the polarization of aging-related research

The second Strategies for Engineered Negligible Senescence conference (SENS II) featured some very provocative ideas. The explicit objective of extending human life span indefinitely has opened a large rift between the meeting's organizer and those who believe he is acting unscientifically, perhaps recklessly. Two SENS conference participants present their views on the divisive nature of SENS.     

基于高效边缘检测的彩色图像灰度化算法

为了使彩色图像灰度化后能更好地保留原始对比度和细节特征，课题组提出一种基于高效边缘检测的彩色图像灰度化算法。课题组通过对线性参数模型产生的一系列备选图像采用像素差值网络（pixel difference networks, PiDiNet）实现鲁棒性、准确性的边缘检测，边界点数最多的备选图像所对应的权重则被选为映射函数的最优解，由于引用了像素差值卷积PDC，对边缘信息敏感度增强。实验结果表明：此算法相比于其他算法能较好地保留原彩色图像的细节特征，使得输出图像轮廓清晰且自然，主客观评价均较优。     

Social Isolation and Spousal Violence: Comparing Female Marriage Migrants With Local Women

This research examined the impact of network participation, social support, and social control on the violence victimization of female marriage migrants by a spouse. Data were from a household survey of 492 cross‐border and 379 local married couples in Hong Kong in 2007. The findings indicated that female marriage migrants were more vulnerable to spousal violence and more socially isolated, compared with local women. Husbands of marriage migrants were also more socially isolated compared with men who married locally. The network participation of both husband and wife was associated positively with spousal violence against women, whereas their perceived social control was associated negatively with violence. Female marriage migrants' greater vulnerability to spousal physical assault was a result of their younger age compared with local women, but their greater vulnerability to spousal psychological aggression and sexual coercion was related, in part, to their own and their husbands' network participation as well as their level of social control.     

In vino veritas

Wine consumption,once predominantly in Europe and the USA, is rapidly evolving into a global trend, with lucrative markets opening up in the Far East, notably Japan, Taiwan, Hong Kong and mainland China.The demand for fine wines, in these regions in particular, sends prices rising through the stratosphere. Lin Y.Hin looks at the professional tasters' ratings and considers issues of score consignment and lifetime estimation.