IV Quantile Regression for Group‐Level Treatments,With an Application to the Distributional Effects of Trade

We present a methodology for estimating the distributional effects of an endogenous treatment that varies at the group level when there are group‐level unobservables, a quantile extension of Hausman and Taylor, 1981. Because of the presence of group‐level unobservables, standard quantile regression techniques are inconsistent in our setting even if the treatment is independent of unobservables. In contrast, our estimation technique is consistent as well as computationally simple, consisting of group‐by‐group quantile regression followed by two‐stage least squares. Using the Bahadur representation of quantile estimators, we derive weak conditions on the growth of the number of observations per group that are sufficient for consistency and asymptotic zero‐mean normality of our estimator. As in Hausman and Taylor, 1981, micro‐level covariates can be used as internal instruments for the endogenous group‐level treatment if they satisfy relevance and exogeneity conditions. Our approach applies to a broad range of settings including labor, public finance, industrial organization, urban economics, and development; we illustrate its usefulness with several such examples. Finally, an empirical application of our estimator finds that low‐wage earners in the United States from 1990 to 2007 were significantly more affected by increased Chinese import competition than high‐wage earners.     

Operational Best Practices in Business Ethics: A Practical and Systematic Benchmarking Tool

This article describes a benchmarking tool managers can use to determine operational policies and processes that could enhance an organization's social and ethical performance. The benchmarking tool consists of a 13‐dimension, 110‐item survey based on an Optimal Ethics Systems Model. These best practices in business ethics are derived from an analysis and assessment of seven institutional infrastructures and accountability standards developed to help managers improve organizational social and ethical performance.     

Uncertainty Distribution Associated with Estimating a Proportion in Microbial Risk Assessment

The uncertainty associated with estimates should be taken into account in quantitative risk assessment. Each input's uncertainty can be characterized through a probabilistic distribution for use under Monte Carlo simulations. In this study, the sampling uncertainty associated with estimating a low proportion on the basis of a small sample size was considered. A common application in microbial risk assessment is the estimation of a prevalence, proportion of contaminated food products, on the basis of few tested units. Three Bayesian approaches (based on beta(0, 0), beta(1/2, 1/2), and beta(l, 1)) and one frequentist approach (based on the frequentist confidence distribution) were compared and evaluated on the basis of simulations. For small samples, we demonstrated some differences between the four tested methods. We concluded that the better method depends on the true proportion of contaminated products, which is by definition unknown in common practice. When no prior information is available, we recommend the beta (1/2, 1/2) prior or the confidence distribution. To illustrate the importance of these differences, the four methods were used in an applied example. We performed two-dimensional Monte Carlo simulations to estimate the proportion of cold smoked salmon packs contaminated by Listeria monocytogenes, one dimension representing within-factory uncertainty, modeled by each of the four studied methods, and the other dimension representing variability between companies.     

Generalized mixed estimator for nonlinear models: a maximum likelihood approach

This paper considers the problem of estimating a nonlinear statistical model subject to stochastic linear constraints among unknown parameters. These constraints represent prior information which originates from a previous estimation of the same model using an alternative database. One feature of this specification allows for the disign matrix of stochastic linear restrictions to be estimated. The mixed regression technique and the maximum likelihood approach are used to derive the estimator for both the model coefficients and the unknown elements of this design matrix. The proposed estimator whose asymptotic properties are studied, contains as a special case the conventional mixed regression estimator based on a fixed design matrix. A new test of compatibility between prior and sample information is also introduced. Thesuggested estimator is tested empirically with both simulated and actual marketing data.     

Sex differentials in premature death among professionals Part II

This is Part II of a two-part article. It explores two hypotheses proposed to explain a reversal of the sex differential in mortality which appears in the 1968–72 death rates of Wisconsin professionals. The first hypothesis proposes that the observed effect is attributable to differentials in the distribution of behavioural risk factors for leading causes of death. Explanatory variables include childlessness, late age at first full-term pregnancy, and relatively high rates of smoking and drinking. Professional men had very low mortality rates from conditions implicating behavioural causes, leading to optimism that low-risk living can introduce a new phase into the epidemiological transition. The second hypothesis posits that the effect may be due to differentials in occupational variables which have systematic gender-divergent outcomes. Occupational levels, tasks, environments and careers all have the potential for such effects. The overall conclusion of the study is that health is systematically related to the quality of support and other conditions in the two major micro-environments for living: work and home.     

General linear processes in Hilbert spaces and prediction

We discuss a general definition of linear processes in Hilbert spaces that takes into account the outstanding role played by this model in prediction theory.     

Asymptotic variances for the multiplicative interaction model

When modelling two-way analysis of variance interactions by a multiplicative term-[Formula] asymptotic variances and covariances are derived for the parameters p, yi and Sj using maximum likelihood theory. The asymptotic framework is defined by a2/K where K is the number of observations per combination of the two factors and a2 the common variance of the eijk values. The results can be applied when K = 1. Two Monte Carlo studies were carried out to check the validity of the formulae for small values of 02/K and to assess their usefulness when replacing the unknown parameters by their estimations. The formulae fit well but the confidence regions produced are too narrow if the interaction term is small. The procedure is illustrated with two examples.     

Agreement Between Mothers and Fathers on the Identification of Infant Communicative Acts: Use of a Randomization Procedure

The extent to which mothers and fathers agree on what they identify as their infant's communicative acts was investigated. Nineteen infants (6 at 6 months, 7 at 9 months, and 6 at 12 months) and their parents participated. A randomization procedure controlled for the frequencies and durations of the communicative acts identified by the parents, and the procedure produced a distribution of 10,000 “chance” agreement values for each parent pair with which their observed level of agreement was compared. The results indicated that, generally, parents could identify their infant's communicative acts consistently, and that observed levels of agreement between parents were significantly higher than would be expected by chance. Differences between mothers and fathers on their identification of communicative acts are considered in terms of the emergence of the infant's intention to communicate.