Screening for prostate cancer using multivariate mixed-effects models

Using several variables known to be related to prostate cancer, a multivariate classification method is developed to predict the onset of clinical prostate cancer. A multivariate mixed-effects model is used to describe longitudinal changes in prostate specific antigen (PSA), a free testosterone index (FTI), and body mass index (BMI) before any clinical evidence of prostate cancer. The patterns of change in these three variables are allowed to vary depending on whether the subject develops prostate cancer or not and the severity of the prostate cancer at diagnosis. An application of Bayes' theorem provides posterior probabilities that we use to predict whether an individual will develop prostate cancer and, if so, whether it is a high-risk or a low-risk cancer. The classification rule is applied sequentially one multivariate observation at a time until the subject is classified as a cancer case or until the last observation has been used. We perform the analyses using each of the three variables individually, combined together in pairs, and all three variables together in one analysis. We compare the classification results among the various analyses and a simulation study demonstrates how the sensitivity of prediction changes with respect to the number and type of variables used in the prediction process.     

Une typologie des économies selon les dimensions de l'emploi dans l'économie informelle

Compte tenu de l'importance des activités informelles, l'auteur propose de classer les pays selon les dimensions de l'emploi dans l'économie informelle plutôt que selon la composition de l'économie formelle. A partir de données du BIT portant sur trente‐six pays en développement, il fait apparaître une corrélation significative entre l'étendue et l'intensité de l'informalité et différents indicateurs sociaux et économiques: PNB par habitant, degré de corruption et de pauvreté et niveau de l'impôt et des cotisations sociales. Les implications de ces résultats pour la réflexion théorique et l'action sont ensuite exposées.     

A comparison of the classical and the linear programming approaches to the classification problem in discriminant analysis

Several mathematical programming approaches to the classification problem in discriminant analysis have recently been introduced. This paper empirically compares these newly introduced classification techniques with Fisher's linear discriminant analysis (FLDA), quadratic discriminant analysis (QDA), logit analysis, and several rank-based procedures for a variety of symmetric and skewed distributions. The percent of correctly classified observations by each procedure in a holdout sample indicate that while under some experimental conditions the linear programming approaches compete well with the classical procedures, overall, however, their performance lags behind that of the classical procedures.     

Improving Risk Assessment Calculations for Traditional Foods Through Collaborative Research with First Nations Communities

As industrial development is increasing near northern Canadian communities, human health risk assessments (HHRA) are conducted to assess the predicted magnitude of impacts of chemical emissions on human health. One exposure pathway assessed for First Nations communities is the consumption of traditional plants, such as muskeg tea (Labrador tea) (Ledum/Rhododendron groenlandicum) and mint (Mentha arvensis). These plants are used to make tea and are not typically consumed in their raw form. Traditional practices were used to harvest muskeg tea leaves and mint leaves by two First Nations communities in northern Alberta, Canada. Under the direction of community elders, community youth collected and dried plants to make tea. Soil, plant, and tea decoction samples were analyzed for inorganic elements using inductively coupled plasma‐mass spectrometry. Concentrations of inorganic elements in the tea decoctions were orders of magnitude lower than in the vegetation (e.g., manganese 0.107 mg/L in tea, 753 mg/kg in leaves). For barium, the practice of assessing ingestion of raw vegetation would have resulted in a hazard quotient (HQ) greater than the benchmark of 0.2. Using measured tea concentrations it was determined that exposure would result in risk estimates orders of magnitude below the HQ benchmark of 0.2 (HQ = 0.0049 and 0.017 for muskeg and mint tea, respectively). An HHRA calculating exposure to tea vegetation through direct ingestion of the leaves may overestimate risk. The results emphasize that food preparation methods must be considered when conducting an HHRA. This study illustrates how collaboration between Western scientists and First Nations communities can add greater clarity to risk assessments.     

Inferences in dichotomous classifications with misclassifications based on sequential repetitive classifications

To create inferences in dichotomous classifications with misclassifications and possibly perform repeated classifications, the maximum likelihood method is commonly used, mainly because of its efficiency in obtaining parameter estimators of a mixture of two binomial distributions. One simpler alternative that is operationally easier is to consider the simple majority method. In this method, each of n items are classified r times as conforming or non-conforming. The final classification of the item is determined by the most frequent class. This method yielded lower mean squared errors than the maximum likelihood and the moments estimators and is asymptotically efficient. In this paper, we introduce a new approach in which the realization of all r repeated classifications of each item may not be needed. Each of n items is sequentially classified as conforming or nonconforming, and the process ceases when the frequency of conforming or non-conforming classification reaches the integer a. We show that, by a Monte Carlo simulation, the last procedure presents a lower mean squared error than the simple majority results for a similar number of r repeated classifications.     

Being of two minds: Ultimatum offers under cognitive constraints

We experimentally investigate how proposers in the Ultimatum Game behave when their cognitive resources are constrained by time pressure and cognitive load. In a dual-system perspective, when proposers are cognitively constrained and thus their deliberative capacity is reduced, their offers are more likely to be influenced by spontaneous affective reactions. We find that under time pressure proposers make higher offers. This increase appears not to be explained by more reliance on an equality heuristic. Analysing the behaviour of the same individual in both roles leads us to favour the strategic over the other-regarding explanation for the observed increase in offers. In contrast, proposers who are under cognitive load do not behave differently from proposers who are not.     

Model bias effects in small area coverage error estimation

For estimation of population totals, dual system estimation (d.s.e.) is often used. Such a procedure is known to suffer from bias under certain conditions. In the following, a simple model is proposed that combines three conditions under which bias of the DSE can result. The conditions relate to response correlation, classification and matching error. The resulting bias is termed model bias. The effects of model bias and synthetic bias in a small area estimation application are illustrated. The illustration uses simulated population data     

Do stepping-stone jobs exist? Early career paths in the medical profession

In the Netherlands, students who want to become a medical specialist have to enrol in a training program which is in limited supply. During the search for a position as trainee (or “junior medical specialist”), they may accept a temporary job as a medical assistant. We use a micro data set to investigate whether such work experience increases the probability of becoming junior medical specialist. To deal with selectivity, we simultaneously model the transitions from unemployment to trainee, from unemployment to medical assistant, from medical assistant to trainee and from medical assistant to unemployment. We find that a job as medical assistant helps to become a medical specialist. Received: 27 July 2000/Accepted: 31 January 2001 All correspondence to Gerard J. van den Berg. Responsible editor: John F. Ermisch.     

Gender effect on housework allocation: Evidence from Spanish two-earner couples

In this paper we study the reasons behind the asymmetric distribution of housework within Spanish two-earner couples. Spouses' housework times are estimated jointly in a bivariate framework using data from the 1991 Work Situation and Time Use Survey. In order to understand the impact of gender-specific factors on the observed allocation of housework, we perform estimations that are in line with the Oaxaca decomposition. Our results suggest that the unequal division of domestic work between wives and husbands in our sample is mainly explained by gender-specific effects rather than by differences in their observable characteristics. All correspondence to Daniel Miles. We thank Juan F. Jimeno, a co-editor and two anonymous referees for helpful comments and the Spanish Instituto de la Mujer for providing us with the data. Financial support from the Cátedra de Estudios Feministas-Caixanova and from the Spanish Dirección General de Ense?anza Superior, grant number BEC2001-1270, is also gratefully acknowledged. Responsible editor: Daniel S. Hamermesh.     

Private versus public financing of education and endogenous growth: A comment on Bräuninger and Vidal

This note shows that the long-run effect in the case of a low skill trap in Br?uninger and Vidal (Journal of Population Economics (2000) 13:387–401) contains a mistake. While not affecting the paper's basic intuition, this implies that the discussion in the short-run analysis also applies in the long-run. Received: 24 April 2001/Accepted: 9 June 2001 I wish to thank Alessandro Cigno and an anonymous referee for useful comments. Responsible editor: Alessandro Cigno.