Bounded influence estimation for regression and scale

Bednarski and Müller [Optimal bounded influence regression and scale M-estimators in the context of experimental design, Statistics 35 (2001), pp. 349–369] introduced a class of bounded influence M estimates for the simultaneous estimation of regression and scale in the linear model with normal errors by solving the corresponding normal location and scale problem at each design point. This limits the proposal to regressor distributions with finite support. Based on their approach, we propose a slightly extended class of M estimates that is not restricted to finite support and is numerically easier to handle. Moreover, we employ the even more general class of asymptotically linear (AL) estimators which, in addition, is not restricted to normal errors. The superiority of AL estimates is demonstrated by numerical comparisons of the maximum asymptotic mean-squared error over infinitesimal contamination neighbourhoods.     

Leaving Home for Marital and Non-marital Reasons in the Netherlands, 1850–1940: The Impact of Parental Death and Parental Remarriage

European Journal of Population - Previous research shows that parentally bereaved children in north-western Europe in the past left home earlier than children who lived together with both...     

On- and offline detection of structural breaks in thermal spraying processes

We investigate and develop methods for structural break detection, considering time series from thermal spraying process monitoring. Since engineers induce technical malfunctions during the processes, the time series exhibit structural breaks at known time points, giving us valuable information to conduct the investigations. First, we consider a recently developed robust online (also real-time) filtering (i.e. smoothing) procedure that comprises a test for local linearity. This test rejects when jumps and trend changes are present, so that it can also be useful to detect such structural breaks online. Second, based on the filtering procedure we develop a robust method for the online detection of ongoing trends. We investigate these two methods as to the online detection of structural breaks by simulations and applications to the time series from the manipulated spraying processes. Third, we consider a recently developed fluctuation test for constant variances that can be applied offline, i.e. after the whole time series has been observed, to control the spraying results. Since this test is not reliable when jumps are present in the time series, we suggest data transformation based on filtering and demonstrate that this transformation makes the test applicable.     

Mortality and economic fluctuations

Using wavelet methods, this paper analyzes the relationship between the age-adjusted, infant, and cause-specific mortality rates and the business cycle in Sweden over the period 1800–2000 (1911–1996 for cause-specific mortality). For the period 1800–2000, an increase in GDP by 1% decreased mortality by 0.7%. This overall relationship is due to a strong counter-cyclical relationship in the nineteenth century, which disappeared in the twentieth century. In contrast, in the twentieth century higher mortality in economic upturns is found for mortality caused by circulatory diseases (including stroke) and accidents.     

Column generation with free replicability in DEA

The evaluation of efficiency scores in data envelopment analysis is based on the construction of artificial decision making units subject to some assumptions, usually requiring convexity of the production possibility set. This demands divisibility in input and output, which is not always possible. The so-called free replicability model, proposed by Henry Tulkens, permits input and output to enter in only discrete amounts. The model is of a mixed integer programming type, for which the number of variables, here corresponding to the decision making units, may be critical in order to reach an optimal solution.     

Emotions and the optimality of uneven tournaments

Standard results from tournament theory show that an employer should avoid organizing uneven tournaments in which workers are characterized by different cost functions. In this paper, we introduce a concept of emotions that workers have when comparing their own performance with the performances of their co-workers. We can show that under emotional workers the employer may prefer uneven tournaments to even ones. In that case, the employer benefits from both positive and negative emotions. Moreover, the behavior of emotional workers fits well with experimental findings on uneven tournaments where players choose excessive effort levels.     

Estimating Consumer Exposure to PFOS and PFOA

Perfluorinated compounds have been used for more than 50 years as process aids, surfactants, and for surface protection. This study is a comprehensive assessment of consumer exposure to perfluorooctane sulfonate (PFOS) and perfluorooctanoic acid (PFOA) from a variety of environmental and product-related sources. To identify relevant pathways leading to consumer exposure to PFOS and PFOA a scenario-based approach has been applied. Scenarios represent realistic situations where age- and gender-specific exposure occurs in the everyday life of consumers. We find that North American and European consumers are likely to experience ubiquitous and long-term uptake doses of PFOS and PFOA in the range of 3 to 220 ng per kg body weight per day (ng/kg_bw/day) and 1 to 130 ng/kg_bw/day, respectively. The greatest portion of the chronic exposure to PFOS and PFOA is likely to result from the intake of contaminated foods, including drinking water. Consumer products cause a minor portion of the consumer exposure to PFOS and PFOA. Of these, it is mainly impregnation sprays, treated carpets in homes, and coated food contact materials that may lead to consumer exposure to PFOS and PFOA. Children tend to experience higher total uptake doses (on a body weight basis) than teenagers and adults because of higher relative uptake via food consumption and hand-to-mouth transfer of chemical from treated carpets and ingestion of dust. The uptake estimates based on scenarios are within the range of values derived from blood serum data by applying a one-compartment pharmacokinetic model.     

Parallelizing computation of expected values in recombinant binomial trees

Recombinant binomial trees are binary trees where each non-leaf node has two child nodes, but adjacent parents share a common child node. Such trees arise in option pricing in finance. For example, an option can be valued by evaluating the expected payoffs with respect to random paths in the tree. The cost to exactly compute expected values over random paths grows exponentially in the depth of the tree, rendering a serial computation of one branch at a time impractical. We propose a parallelization method that transforms the calculation of the expected value into an embarrassingly parallel problem by mapping the branches of the binomial tree to the processes in a multiprocessor computing environment. We also discuss a parallel Monte Carlo method and verify the convergence and the variance reduction behavior by simulation study. Performance results from R and Julia implementations are compared on a distributed computing cluster.