Absolute penalty and shrinkage estimation in partially linear models

In the context of a partially linear regression model, shrinkage semiparametric estimation is considered based on the Stein-rule. In this framework, the coefficient vector is partitioned into two sub-vectors: the first sub-vector gives the coefficients of interest, i.e., main effects (for example, treatment effects), and the second sub-vector is for variables that may or may not need to be controlled. When estimating the first sub-vector, the best estimate may be obtained using either the full model that includes both sub-vectors, or the reduced model which leaves out the second sub-vector. It is demonstrated that shrinkage estimators which combine two semiparametric estimators computed for the full model and the reduced model outperform the semiparametric estimator for the full model. Using the semiparametric estimate for the reduced model is best when the second sub-vector is the null vector, but this estimator suffers seriously from bias otherwise. The relative dominance picture of suggested estimators is investigated. In particular, suitability of estimating the nonparametric component based on the B-spline basis function is explored. Further, the performance of the proposed estimators is compared with an absolute penalty estimator through Monte Carlo simulation. Lasso and adaptive lasso were implemented for simultaneous model selection and parameter estimation. A real data example is given to compare the proposed estimators with lasso and adaptive lasso estimators.     

HIGH FREQUENCY MEASUREMENTS OF CUTTING FORCES IN MILLING BY INVERSE FILTERING

Accurate estimates of cutting forces are important in the evaluation of different cutting tool geometries and concepts. However, dynamic influences from the measurement system affect the result, which can make the obtained cutting force data erroneous and misleading. This article presents a method to obtain an inverse filter which compensates for the dynamic influences of the measurement system. Using this approach, unwanted dynamic effects of the measurement system can be counteracted, making it possible to retain information related to the cutting forces contained in the high frequency region. The advantage of the proposed method is illustrated by comparing simulated, inverse- and low-pass filtered forces to unfiltered forces under different cutting conditions. The results show that inverse filtering increases the usable frequency range of the force dynamometer and thereby provide more reliable results compared to both low-pass and unfiltered forces.     

On the influence of Pt particle size on the PEMFC cathode performance

Colloidal suspensions of almost spherical and crystalline Pt nanoparticles between 1.6 and 2.6 nm in diameter and with narrow size distribution were synthesized using the phase transfer method (PTM) with alkylamines, C_nNH₂, as stabilizing agents. Batches of such homogenous Pt-C_nNH₂ (n = 8, 12) nanocrystals were deposited onto Vulcan XC-72 carbon powder, and the activity for the oxygen reduction reaction (ORR) of this series of Pt/C materials was evaluated under PEMFC conditions. The aim was to elucidate whether this type of stabilized Pt nanoparticles were as active for the ORR as a corresponding commercial Pt/C material, and if any difference in mass activity could be observed between catalysts with different Pt particle size. In the PEMFC experiments, i.e. voltammetry in oxygen and nitrogen, it was found that, after an initial electrode activation, the ORR activity of the catalysts prepared from the alkylamine-stabilized Pt nanoparticles deposited on carbon was as high as that of the employed commercial reference catalyst. In fact, all samples in the Pt/C series showed high and very similar ORR activity normalized to Pt-loading, without significant dependence on the initial Pt particle size. However, pre- and post-electrochemical characterization of the Pt/C material series with TEM showed that structural changes of the Pt nanoparticles occurred during electrochemical evaluation. In all samples studied the mean Pt particle size increased during the electrochemical evaluation resulting in decreased differences between the samples explaining the observed similar ORR performance of the different materials. These results emphasize the necessity of post-operation characterization of fuel cell catalysts when discussing electrocatalytic activity. In addition, employing complex preparation efforts for lowering the Pt particle size below 3 nm may have limited practical value unless the particles are stabilized from electrochemical sintering.     

Experiences from experimental and numerical springback studies of a semi-industrial forming tool

The work reported in the present paper constitutes a part of a project on simulation of springback in sheet metal forming. Previous work in this project has been concentrated on material modeling and characterization with focus on springback applications. It has been demonstrated that, with proper considerations of all aspects of the material model and the material properties, excellent springback results can be obtained for simple problems. At the simulation of real, industrial parts, a number of additional problems are encountered. Many of these problems are associated with deviations from nominal geometries and other properties. These are examples of factors that influence the outcome of the forming process, but are unknown to the analyst, and can therefore not be considered in the simulation of the forming process. Other phenomena are known to exist, but due to their complexity, they are practically impossible to consider in industrial simulations. Examples of such phenomena are the true frictional behavior in contacts between the blank and the tools, and the flexibility of the press and the forming tool. The influence of these kinds of effects is discussed in the present paper. In the current study, a semi-industrial tool, specially designed to catch those springback problems that are encountered in the forming of real industrial, parts, is used. The tool includes several characteristics that can be found in typical forming tools, such as several draw radius steps and change-over in section geometries. Effects like flange/wall angle changes, sidewall curl and twist are obtained at springback. The sensitivity of the predicted springback is evaluated with respect to various numerical factors, such as the friction coefficient, the material model, and the mesh density. Finally, the quality of the predicted springback behavior for four different materials, commonly used in the automotive industry, is evaluated.     

Young drivers--reduced crash risk with passengers in the vehicle

Studies have shown that the effect of passengers on accident propensity among young drivers is ambiguous--in some cases positive and some negative. In Sweden, various kinds of information are compiled in registers, including a national accident database and exposure data collected in a national investigation of the driving habits of license holders. Access to such data offers a good opportunity to study crash risks related to driving with and without passengers. This was done for drivers in three different age groups (18-24, 25-64 and >65 years) accompanied by one, two or three or more passengers. Differences in crash risk were estimated using incidence density ratios (IDRs) and 95% confidence intervals. The results show that passengers had an overall protective effect, that is, the crash risk was higher for those who drove alone, regardless of their age or gender. This protective effect increased with every extra passenger (up to eight), indicating that the more passengers in the vehicle, the safer the driving. The influence of passengers was weakest (albeit still positive) among the youngest drivers (ages 18-24 years), especially the males in that group. The protective impact showed the same pattern on all days of the week, but was most marked from Friday to Sunday for most of the drivers, regardless of age.     

Water pressure chamber for static testing of panels

An experimental device has been developed for the quasi-static testing of panels exposed to equally distributed pressure loading. This test equipment uses water as the pressure medium and may be operated in several ways. The test facility is described and various operational aspects are discussed. Emphasis has been put on similarity in experimental test conditions and numerical modelling in order to avoid uncertainties that are not deducible from geometrical measurements or materials testing. Examples are presented of tests that have been performed using this water pressure chamber. The experiments have been performed to obtain validation tests for thin-walled corrugated steel panels as well as concrete slabs.     

Understanding the role of personality in explaining associations between marital quality and parenting.

Analyses assessed the degrees to which personality accounts for associations between marital quality and parenting and mediates genetic contributions to these relationships. Participants included 318 male and 544 female same-sex twin pairs from the Twin and Offspring Study in Sweden. All twins completed self-report measures of marital quality and personality (anxiousness, aggression, sociability). Composite measures of parent negativity and warmth were derived from the twins’ and their adolescent children’s ratings of the twins’ disciplinary styles and the emotional tone of the parent–child relationship. Observational ratings of marital quality and parenting were also obtained for a subset of twins. Personality characteristics explained 33% to 42% of the covariance between reported marital quality and parenting and 26% to 28% of the covariance between observed marital quality and parenting. For both sets of analyses, personality accounted for more than half of the genetic contributions to covariance between marital quality and parenting. Results indicate that personality significantly contributes to associations between marital quality and parenting and that personality is an important path through which genetic factors contribute to family relationships. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Dinoflagellate cysts and hydrographical change in Gullmar Fjord, west coast of Sweden

This high-resolution study of the latest Holocene dinoflagellate cyst record from Gullmar Fjord, on the west coast of Sweden, provides evidence for the recognition of two major dinoflagellate communities within the fjord over the last 85 years. These communities may have their origins with the history of cultural eutrophication within the region, but are more likely to be associated with the wider phenomenon of the North Atlantic Oscillation and/or the complex hydrographical response of the fjord to various changing climatic environments between 1915 and 1999. The changing dinoflagellate cyst populations are compared in detail with the many hydrographical parameters available from this well studied fjord with its long instrumental records. Indeed the dinoflagellate cysts fail to demonstrate a convincing ongoing eutrophication record for the fjord but do show a major change in the cyst assemblages at about 1969/1970 at a time when the NAO was changing from a negative phase to the present-day positive phase. Gullmar Fjord is important in the history of dinoflagellate cyst studies, being the site of the 1954 study by Erdtman in which viable cysts, produced within the phytoplankton, were first documented within the water column.