Predicting Corporate Bankruptcy: An Evaluation of Alternative Statistical Frameworks

Corporate bankruptcy prediction has attracted significant research attention from business academics, regulators and financial economists over the past five decades. However, much of this literature has relied on quite simplistic classifiers such as logistic regression and linear discriminant analysis (LDA). Based on a large sample of US corporate bankruptcies, we examine the predictive performance of 16 classifiers, ranging from the most restrictive classifiers (such as logit, probit and linear discriminant analysis) to more advanced techniques such as neural networks, support vector machines (SVMs) and “new age” statistical learning models including generalised boosting, AdaBoost and random forests. Consistent with the findings of Jones et al. ( 2015 ), we show that quite simple classifiers such as logit and LDA perform reasonably well in bankruptcy prediction. However, we recommend the use of “new age” classifiers in corporate bankruptcy modelling because: (1) they predict significantly better than all other classifiers on both the cross‐sectional and longitudinal test samples; (2) the models may have considerable practical appeal because they are relatively easy to estimate and implement (for instance, they require minimal researcher intervention for data preparation, variable selection and model architecture specification); and (3) while the underlying model structures can be very complex, we demonstrate that “new age” classifiers have a reasonably good level of interpretability through such metrics as relative variable importances (RVIs).     

Sample attrition in the RLMS, 2001–10

The data of the Russian Longitudinal Monitoring Survey (RLMS) – Higher School of Economics represents one of the few nationally representative sources of household and individual data for Russia. These data have been collected since 1992 and in recent years, thanks to more secure financial and logistical support, have become a resource increasingly drawn upon by scholars and students for national and cross-national studies. In this paper, we examine the extent of non-random attrition in the RLMS and discuss the circumstances under which this might give rise to biases in econometric analysis. We illustrate this with an example drawn from the health sphere.     

Up in smoke: demarketing and consumer smoking cessation

The concept of demarketing refers to the use of marketing techniques to reduce or eliminate demand for a product or service. A review of the demarketing literature relating to health and specifically antismoking initiatives indicated that, while research on this topic exists, much of it is not grounded in an acceptable attitudinal or behavioural theory. After determining the importance placed by a sample of 18–24‐year olds on nine demarketing initiatives, two dimensions were identified that best explained this construct. Items within these dimensions were summed and averaged to form single variables, which were then used to form the attitudinal component of the Model of Goal Directed Behaviour. The findings showed that two of these variables – one that captured product packaging aspects and another that consisted of place and price items – significantly influenced the desire to quit and indirectly influenced the intention to quit. Anticipated positive emotions, frequency of quitting attempts and perceived control over quitting also positively influenced the desire and/or intention to quit. The article concludes with a discussion that interprets these findings from a theoretical and practical perspective and suggests directions for future research.     

Standards,learning, and growth in Britain, 1901–2009

This article considers the model of voluntary, consensus based standardization as developed through the British Standards Institution (BSI) and its contribution to learning and productivity growth. It discusses the significant role played by professional engineers in the model's introduction, its extension at home, and its imitation overseas. It is argued that by 1931 the BSI catalogue of standards represented a considerable stock of codified knowledge whose growth reflected underlying aggregate technological opportunities, assisting in their transformation into technological advance. To help validate this claim, a measure of the size of the BSI catalogue of standards is incorporated into an econometric model of aggregate productivity growth in Britain. Findings show that the growth of the standards catalogue is associated with a substantial proportion of labour productivity growth over the period 1931–2009. Estimates relating to the short‐run dynamics involved are consistent with the idea that there are causal linkages running from standards to growth. When interpreting these findings, it is argued that the overall weight of historical evidence points to standardization—coordinated through the BSI—as providing an important path of learning for the British economy over the period considered.     

Panel Data Analysis of Regional Differentials in the Registered Voter Turnout Rate and the Expected Benefits of Voting for Minorities

Voter participation rates vary widely across the 50 states. We seek to identify a source of this interstate variation, focusing upon the 2000, 2004, 2008, and 2012 general elections. Of note is that the latter two general elections featured a minority (i.e., African-American) nominee for president from one of the major political parties. Within the context of the rational voter model, we hypothesize that the presence of a minority candidate atop the ticket will boost minorities’ expected net benefits from voting, with the result that minority voter participation will be higher. Based on cross-sectional fixed-effects estimations, we find that states’ voter participation rates were unaffected by the percent of the population that was either African-American or Hispanic for the 2000 and 2004 election cycles, while states’ voter participation rates were positively related to the percent of the population that was either African-American or Hispanic for the 2008 and 2012 election cycles.     

Health insurer market power and primary care consolidation

This paper analyzes how health insurance market concentration impacts the market structure of primary care physicians. In more concentrated insurance markets, physicians are found to work in larger practices and their practices are more likely to have a hospital with an ownership interest. Physicians are also less likely to report being in a competitive physician market, consistent with practice consolidation. Our results suggest that consolidation in insurance markets impacts the competitive structure of physician markets.     

From Supply Chains to Supply Ecosystems: Implications for Strategic Sourcing Research and Practice

Traditionally, sourcing decisions such as what to buy and from whom to buy it were viewed as clerical in nature. Today, however, many executives view sourcing decisions as strategic and these decisions are being made from more prominent positions within firms. Indeed, over time more companies are embracing the concept of strategic sourcing—making acquisition decisions with the intent of creating value and achieving a competitive advantage. In this article, we introduce the Special Topic Forum articles on strategic sourcing and performance. Each offers important implications for strategic sourcing research and practice. An overall implication is that some traditional supply chains appear to be giving way to supply ecosystems, which represent a set of interdependent and coordinated organizations that share common adaptive challenges and that collectively shape the creation and nurturing of a sourcing base that contributes to competitive advantage and superior performance. We offer several theoretical perspectives that hold promise for explaining supply ecosystem functioning and outcomes. In terms of practice, firms need to adapt their strategic sourcing to this evolution or run the risk of being weakened.     

Ghanaian school-going adolescents’ self-perceived barriers of access to quality education: A qualitative study

Social equality has emerged as one of the important ideals of human society, especially for development. Moreover, education has featured prominently among the many key drivers of equality. However, in developing nations like Ghana, education represents one of the areas in which gross inequality has eroded the country’s potential for attaining equality and development. The aim of the current study was to explore adolescents’ self-perceived barriers regarding access to quality education in the Northern Region of Ghana and provide suggestions on the way forward. A sample of 18 school-going adolescents from six secondary schools within the Northern Region of Ghana was purposively selected to participate in face-to-face interviews. The barriers identified included limited resources, gender disparities and a multi-dimension nature of poverty. Implications for policy and research are discussed.