A Brief Survey of Modern Optimization for Statisticians

Modern computational statistics is turning more and more to high‐dimensional optimization to handle the deluge of big data. Once a model is formulated, its parameters can be estimated by optimization. Because model parsimony is important, models routinely include non‐differentiable penalty terms such as the lasso. This sober reality complicates minimization and maximization. Our broad survey stresses a few important principles in algorithm design. Rather than view these principles in isolation, it is more productive to mix and match them. A few well‐chosen examples illustrate this point. Algorithm derivation is also emphasized, and theory is downplayed, particularly the abstractions of the convex calculus. Thus, our survey should be useful and accessible to a broad audience.     

Successful programs wanted: Exploring the impact of alignment

Alignment between formulation and implementation of business strategy can be important for achieving successful programs. The authors have explored the development of a program management alignment theory. Statistical testing showed that interaction between the study model variables was found to be multidimensional, complex, and subtle in influence. Thus, the authors conclude that programs have both deliberate and emergent strategies requiring design and management to be organized as complex adaptive systems. Program life‐cycle phases of design and transition were often formed from an unclear and confusing strategic picture at the outset, which can make those phases difficult to control. Learning was established as an underlying challenge. The study model demonstrated continuous alignment as an essential attribute contributing toward successful delivery. This requires program design and structure to adopt an adaptive posture.     

Boardroom centrality and firm performance

Firms with central boards of directors earn superior risk-adjusted stock returns. A long (short) position in the most (least) central firms earns average annual returns of 4.68%. Firms with central boards also experience higher future return-on-assets growth and more positive analyst forecast errors. Return prediction, return-on-assets growth, and analyst errors are concentrated among high growth opportunity firms or firms confronting adverse circumstances, consistent with boardroom connections mattering most for firms standing to benefit most from information and resources exchanged through boardroom networks. Overall, our results suggest that director networks provide economic benefits that are not immediately reflected in stock prices.     

The Structure of the American Economy

We explore the relationship between input–output accounts and the national revenue function. The generalized inverse of an economy's technology matrix carries information relating changes in endowments with changes in outputs; its transpose relates output prices and factor prices. Our primary theoretical contribution is to derive an economy's revenue function for an arbitrary Leontief technology. Our main empirical contribution is to compute the national revenue function for the American economy in 2003 and to describe its properties. We implement our ideas using two different models: one where all factors are mobile and another with sector‐specific capital.     

Underwriting the elderly: the utility of DWR, part II

Cognitive function is an important predictor of mortality in the elderly. Over the past few years, an increasing number of life insurance companies have incorporated screening for cognitive impairment into the underwriting process at older ages. Many different test instruments provide a measurement of cognitive function. Among these, the 10 word delayed word recall test (DWR) offers the best opportunity to study mortality directly, because of a long history of use in long-term care risk assessment. This article revises and extends a previous report published in the Journal in 2006 looking at the relationship between DWR score and mortality.     

Identification and information in monotone binary models

This paper considers binary response models where errors are uncorrelated with a set of instrumental variables and are independent of a continuous regressor v$v$, conditional on all other variables. It is shown that these exclusion restrictions are not sufficient for identification and that additional identifying assumptions are needed. Such an assumption, introduced by Lewbel [Semiparametric qualitative response model estimation with unknown heteroskedasticity or instrumental variables. Journal of Econometrics 97, 145–177], is that the support of the continuous regressor is large, but we show that it significantly restricts the class of binary phenomena which can be analysed. We propose an alternative additional assumption under which β$\beta$ remains just identified and the estimation unchanged. This alternative assumption does not impose specific restrictions on the data, which broadens the scope of the estimation method in empirical work. The semiparametric efficiency bound of the model is also established and an existing estimator is shown to achieve that bound. The efficient estimator uses a plug-in density estimate. It is shown that plugging in the true density rather than an estimate is inefficient. Extensions to ordered choice models are provided.     

Health care reform: universal access is feasible and necessary

While other industrial nations' health care systems have their own problems, they have more leeway to address those problems than does the United States, which spends twice as much on per capita health care as the average for other industrial capitalist democracies yet ranks average or below average in many comparative measures of health care quality. In fact, the authors of this article argue that international experience shows that assurance of universal access through expanded government involvement could provide savings while actually improving the quality of U.S. health care. In addition, universal access would recognize health care as a basic human right, not a commodity to be bartered in the marketplace and allocated based on class, race and social position.     

On the importance of matching strategic behavior and target market selection to business strategy in high-tech markets

Business strategy is fundamentally concerned with the actions required to create superior customer value in the firm’s target markets with the ultimate goal of achieving superior performance. Marketing theory suggests that two critical marketing activities required to achieve this end are: (1) the adoption of appropriate strategic behaviors (i.e., customer-oriented, competitor-oriented, technology-oriented) and (2) targeting of the appropriate market segments (i.e., innovators, early adopters, early majority, late majority, laggards). This study builds on prior research which demonstrates that the strategic behavior—firm performance relationship is contingent on the firm’s strategy by examining this relationship in high tech markets and by considering the incremental contribution of appropriate target market selection. Responses from 160 senior marketing managers in high-tech firms reveal strong support for our framework. Thus, this study provides useful guidance to executives and managers in high-tech firms regarding the steps that they should take to increase their probability of success.