A spanning heuristic for the unordered and ordered matching identification problem

The matching identification problem (MIP) is a combinatoric search problem related to the fields of learning from examples, boolean functions, and knowledge acquisition. The MIP involves identifying a single “goal” item from a large set of items. Because there is commonly a cost associated with evaluating each guess, the goal item should be identified in as few guesses as possible. As in most search problems, the items have a similar structure, which allows an evaluation of each guessed item. In other words, each guessed item elicits partial information about the goal item, i.e. how similar the guess is to the goal. With this information the goal is more quickly identified.The unordered MIP has been studied by Mehrez and Steinberg (ORSA J. Comput. 7 (1995) 211) in which they proposed two different types of algorithms. The purpose of the present paper is to suggest an improved Spanning Heuristic algorithm. Its improvement increases as the problem size increases. Further results and comparisons are derived for the unordered and ordered cases.This research shows that when the search space is very large, it is better to inquire from items that are known not to be the goal (they have been ruled out by previous guesses), for the purpose of acquiring more information about the goal. As the search space is narrowed, it is better to guess items that have not been ruled out.     

Leveraging the common and outsourcing the distinct: institutional difference and multinational company identity in emerging economies

ABSTRACT

This paper explores the challenges of outsidership faced by multinational companies (MNCs) accessing emerging economies. The paper departs from popular conceptualisations of culture drawing instead on social identity theory as part of an institutions-based view, to explain how employing a singularities approach can allow the identification of commonalities of Indianness and Australianness that can be leveraged to ameliorate negative socially embedded business norms that pose a challenge to MNC success in emerging economies. This study shows that despite recognised formal (regulatory) and informal (cultural) differences, managers of Australian MNCs operating in India were able to innovatively identify and exploit cultural commonalties/singularities to ease cultural adjustment and mitigate intergroup conflict, allowing them to achieve an innovative institutional fit. The paper discusses how, using this approach, sole-venture MNCs can access many of the benefits provided by collaborative entry modes while preserving MNC identity. We further propose that the singularities approach can provide a valuable starting point to develop culturally appropriate competence training initiatives.     

Model-based clustering with sparse covariance matrices

Statistics and Computing - Finite Gaussian mixture models are widely used for model-based clustering of continuous data. Nevertheless, since the number of model parameters scales quadratically with...     

Promoting Fathers' Engagement With Children: Preventive Interventions for Low‐Income Families

Few programs to enhance fathers' engagement with children have been systematically evaluated, especially for low‐income minority populations. In this study, 289 couples from primarily low‐income Mexican American and European American families were randomly assigned to one of three conditions and followed for 18 months: 16‐week groups for fathers, 16‐week groups for couples, or a 1‐time informational meeting. Compared with families in the low‐dose comparison condition, intervention families showed positive effects on fathers' engagement with their children, couple relationship quality, and children's problem behaviors. Participants in couples' groups showed more consistent, longer term positive effects than those in fathers‐only groups. Intervention effects were similar across family structures, income levels, and ethnicities. Implications of the results for current family policy debates are discussed.     

The Sociology of Inequality and the Rise of Neo-Inequality

Sociologists’ conventional approaches to understanding inequality and poverty do not adequately capture the causes, extent, and consequences of new forms of inequality that have emerged in the late 20th and early 21st centuries. Following a critique of conventional sociological theorizing on inequality and stratification, a synthesis of new sociological/social psychology thinking on inequality, and the use of case study data from Detroit, I explore, describe, and explain how neo-liberal ideology, neo-liberal systems of governance, and neo-liberal policies have created the neo-inequalities that now define our age.     

Best practice in couple relationship education

Relationship education is widely available to couples and is intended to reduce the prevalence of relationship distress, divorce, and the associated personal and social costs. To realize the potential benefits of couple relationship education, it needs to be evidence-based, offered in ways that attract couples at high-risk for relationship problems, and focused on factors that put couples at high-risk for future relationship problems.     

Discovering classes in microarray data using island counts

We present a new biclustering algorithm to simultaneously discover tissue classes and identify a set of genes that well-characterize these classes from DNA microarray data sets. We employ a combinatorial optimization approach where the object is to simultaneously identify an interesting set of genes and a partition of the array samples that optimizes a certain score based on a novel color island statistic. While this optimization problem is NP-complete in general, we are effectively able to solve problems of interest to optimality using a branch-and-bound algorithm. We have tested the algorithm on a 30 sample Cutaneous T-cell Lymphoma data set; it was able to almost perfectly discriminate short-term survivors from long-term survivors and normal controls. Another useful feature of our method is that can easily handle missing expression data.     

How Well-Informed Are Pension Scheme Members on Their Future Pension Benefits? Evidence from Ireland

One part of the policy response in many countries to increasing pension coverage will be greater private provision on the part of individuals. This requires that individuals are well informed about pensions. In this article, we assess levels of knowledge of pensions using a representative sample of older Irish adults. We find that two-thirds of individuals enrolled in pension schemes do not know what amount will be paid out on retirement and/or whether the payments will be in the form of lump sums, monthly payments, or both. One policy implication is the need for increased information to be directed at certain groups, in particular, women and less educated people. More fundamentally, the results suggest that the mandatory elements in pension systems should be extended.     

Mixture of latent trait analyzers for model-based clustering of categorical data

Model-based clustering methods for continuous data are well established and commonly used in a wide range of applications. However, model-based clustering methods for categorical data are less standard. Latent class analysis is a commonly used method for model-based clustering of binary data and/or categorical data, but due to an assumed local independence structure there may not be a correspondence between the estimated latent classes and groups in the population of interest. The mixture of latent trait analyzers model extends latent class analysis by assuming a model for the categorical response variables that depends on both a categorical latent class and a continuous latent trait variable; the discrete latent class accommodates group structure and the continuous latent trait accommodates dependence within these groups. Fitting the mixture of latent trait analyzers model is potentially difficult because the likelihood function involves an integral that cannot be evaluated analytically. We develop a variational approach for fitting the mixture of latent trait models and this provides an efficient model fitting strategy. The mixture of latent trait analyzers model is demonstrated on the analysis of data from the National Long Term Care Survey (NLTCS) and voting in the U.S. Congress. The model is shown to yield intuitive clustering results and it gives a much better fit than either latent class analysis or latent trait analysis alone.