Oracle Inequalities for Convex Loss Functions with Nonlinear Targets

This article considers penalized empirical loss minimization of convex loss functions with unknown target functions. Using the elastic net penalty, of which the Least Absolute Shrinkage and Selection Operator (Lasso) is a special case, we establish a finite sample oracle inequality which bounds the loss of our estimator from above with high probability. If the unknown target is linear, this inequality also provides an upper bound of the estimation error of the estimated parameter vector. Next, we use the non-asymptotic results to show that the excess loss of our estimator is asymptotically of the same order as that of the oracle. If the target is linear, we give sufficient conditions for consistency of the estimated parameter vector. We briefly discuss how a thresholded version of our estimator can be used to perform consistent variable selection. We give two examples of loss functions covered by our framework.     

Integrating Poverty and Environmental Concerns into Value‐Chain Analysis: A Strategic Framework and Practical Guide

This article aims to guide the design and implementation of action‐research projects in value‐chain analysis by presenting a strategic framework focused on small producers and trading and processing firms in developing countries. Its stepwise approach – building on the conceptual framework set out in a companion article – covers in detail what to do, questions to be asked and issues to be considered, and integrates poverty, gender, labour and environmental concerns.‘Upgrading’ strategies potentially available for improving value‐chain participation for small producers are identified, with the ultimate purpose of increasing the rewards and/or reducing the risks.     

Sharp Threshold Detection Based on Sup-Norm Error Rates in High-Dimensional Models

We propose a new estimator, the thresholded scaled Lasso, in high-dimensional threshold regressions. First, we establish an upper bound on the ?_∞ estimation error of the scaled Lasso estimator of Lee, Seo, and Shin. This is a nontrivial task as the literature on high-dimensional models has focused almost exclusively on ?₁ and ?₂ estimation errors. We show that this sup-norm bound can be used to distinguish between zero and nonzero coefficients at a much finer scale than would have been possible using classical oracle inequalities. Thus, our sup-norm bound is tailored to consistent variable selection via thresholding. Our simulations show that thresholding the scaled Lasso yields substantial improvements in terms of variable selection. Finally, we use our estimator to shed further empirical light on the long-running debate on the relationship between the level of debt (public and private) and GDP growth. Supplementary materials for this article are available online.     

Generalized ratio-cum-product type exponential estimator in stratified random sampling

ABSTRACT

In this article, we propose a generalized ratio-cum-product type exponential estimator for estimating population mean in stratified random sampling. Asymptotic expression of the bias and mean squared error of the proposed estimator are obtained. Asymptotic optimum estimator in the proposed estimator has been obtained with its mean squared error formula. Conditions under which the proposed estimator is more efficient than usual unbiased estimator, combined ratio and product type estimators, Singh et al. (2008 Singh, R., Kumar, M., Singh, R.D., Chaudhary, M.K. (2008). Exponential ratio type estimators in stratified random sampling. Presented in International Symposium on Optimisation and Statistics (I.S.O.S) at A.M.U., Dec. 2008, 29–31, Aligarh, India. [Google Scholar]) estimators and Tailor and Chouhan (2014 Tailor, R., Chouhan, S. (2014). Ratio-cum-product type exponential estimator of finite population mean in stratified random sampling. Commun. Statist. Theor. Meth. 43:343–354.[Taylor & Francis Online], [Web of Science ®] , [Google Scholar]) estimator are obtained. An empirical study has also been carried out.     

Work & stress: Information

The aim of this study was to investigate whether psychosocial factors at work are related to low-back pain among nursing personnel. It was conducted as a short-term follow-up study of 200 Danish female nursing personnel, providing care for the elderly. Self-reported measures of psychosocial factors such as time pressure, emotional demands of clients, control and social support were obtained by questionnaire at baseline, while stress, physical exertion and low-back pain were reported by diary records made in two, three-day periods in the subsequent 6 months. The study examines the possible influence of psychosocial factors preceding low-back pain as well as stress and physical exertion reported simultaneously with low-back pain. Only stress was associated with low-back pain (Odds Ratio (OR)=2.3; Confidence Interval (CI)=1.3-3.9) while neither physical exertion or any of the psychosocial factors were related to low-back pain. A possible pathway connecting emotional demands of clients to low-back pain through the mediation of stress was suggested.     

Are psychosocial factors associated with low-back pain among nursing personnel?

The aim of this study was to investigate whether psychosocial factors at work are related to low-back pain among nursing personnel. It was conducted as a short-term follow-up study of 200 Danish female nursing personnel, providing care for the elderly. Self-reported measures of psychosocial factors such as time pressure, emotional demands of clients, control and social support were obtained by questionnaire at baseline, while stress, physical exertion and low-back pain were reported by diary records made in two, three-day periods in the subsequent 6 months. The study examines the possible influence of psychosocial factors preceding low-back pain as well as stress and physical exertion reported simultaneously with low-back pain. Only stress was associated with low-back pain (Odds Ratio (OR)=2.3; Confidence Interval (CI)=1.3-3.9) while neither physical exertion or any of the psychosocial factors were related to low-back pain. A possible pathway connecting emotional demands of clients to low-back pain through the mediation of stress was suggested.     

Communicating about the Risks and Benefits of Genetically Modified Foods: The Mediating Role of Trust

Recent research suggests that public attitudes toward emerging technologies are mainly driven by trust in the institutions promoting and regulating these technologies. Alternative views maintain that trust should be seen as a consequence rather than a cause of such attitudes. To test its actual role, direct as well as mediating effects of trust were tested in an attitude change experiment involving 1,405 consumers from Denmark, Germany, Italy, and the United Kingdom. After prior attitudes to genetic modification in food production had been assessed, participants received different information materials (either product-specific information or balanced/general information about genetic modification in food production) and were asked to evaluate different types of genetically modified foods (either beer or yoghurt). The information materials were attributed to different information sources (either an industry association, a consumer organization, or a government source). After completion, perceived risk and perceived benefit were assessed, and participants indicated their trust in the information sources to which the materials had been attributed. Direct and trust-mediated attitude change effects were estimated in a multi-sample structural equation model. The results showed that information provision had little effect on people's attitudes toward genetically modified foods, and that perceptions of information source characteristics contributed very little to attitude change. Furthermore, the type of information strategy adopted had almost no impact on postexperimental attitudes. The extent to which people trusted the information sources appeared to be driven by people's attitudes to genetically modified foods, rather than trust influencing the way that people reacted to the information. Trust was not driving risk perception-rather, attitudes were informing perceptions of the motivation of the source providing the information.