Subsampling tests of parameter hypotheses and overidentifying restrictions with possible failure of identification

We introduce a general testing procedure in models with possible identification failure that has exact asymptotic rejection probability under the null hypothesis. The procedure is widely applicable and in this paper we apply it to tests of arbitrary linear parameter hypotheses as well as to tests of overidentification in time series models given by unconditional moment conditions. The main idea is to subsample classical tests, like for example the Wald or the J test. More precisely, instead of using critical values based on asymptotic theory, we compute data-dependent critical values based on the subsampling technique.We show that under full identification the resulting tests are consistent against fixed alternatives and that they have exact asymptotic rejection probabilities under the null hypothesis independent of identification failure. Furthermore, the subsampling tests of parameter hypotheses are shown to have the same local power as the original tests under full identification.An algorithm is provided that automates the block size choice needed to implement the subsampling testing procedure. A Monte Carlo study shows that the tests have reasonable size properties and often outperform other robust tests in terms of power.     

A large-scale application of the partial coverage uncapacitated facility location problem

The traditional, uncapacitated facility location problem (UFLP) seeks to determine a set of warehouses to open such that all retail stores are serviced by a warehouse and the sum of the fixed costs of opening and operating the warehouses and the variable costs of supplying the retail stores from the opened warehouses is minimized. In this paper, we discuss the partial coverage uncapacitated facility location problem (PCUFLP) as a generalization of the uncapacitated facility location problem in which not all the retail stores must be satisfied by a warehouse. Erlenkotter's dual-ascent algorithm, DUALOC, will be used to solve optimally large (1600 stores and 13?000 candidate warehouses) real-world implemented PCUFLP applications in less than two minutes on a 500?MHz PC. Furthermore, a simple analysis of the problem input data will indicate why and when efficient solutions to large PCUFLPs can be expected.     

Error-Controlled Model Order Reduction of Second Order Systems with Adaptive Choice of Shifts and Order

This contribution summarizes a novel approach to error-controlled model order reduction of second order systems, i. e. a new way to find a reduced order model of specified approximation quality. To this end, starting from a particular state space realization of the model, one performs multiple reduction steps and increments the reduced model until (computationally affordable) rigorous global error bounds fall below the prescribed limit. (© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim)     

Multiplicity distribution inK + Al andK + Au collisions at 250 GeV/c and a test of the multiple collision model

Multiplicity distributions, observed inK ⁺ interactions with Al and Au nuclei at 250 GeV/c incident momentum are presented. They are analyzed in the framework of multiple collisions of the incident particle inside a nucleus. The probability distribution of the number of grey tracks is well described by the model of Andersson et al., if a negative binomial distribution is assumed for the distribution of the number of grey protons produced per elementary collision instead of the usual geometrical distribution. The analysis of the average and dispersion of the charge multiplicity distribution supports the validity of the multiple collision model, including results on correlations between forward and backward multiplicities.     

Development of a quantitative FI-MS technique for the characterization of high- and nonboiling saturated hydrocarbon mixtures

Summary Using a saturated non-boiling hydrocarbon mixture, the influence of two parameters on the results of field ionization mass spectrometry (FI-MS) measurements was studied: (a) the potential difference between the FI emitter and the counterelectrode; (b) the emitter temperature.Variation of the potential difference had only a minor effect on the average molecular mass measured and had no evident effect on the relative ring number distribution in the sample. In contrast, when the emitter temperature was increased, higher average molecular masses were recorded. Moreover, the average molecular masses shifted to higher ring numbers. In order to control the relationship between the described influences during mixture analysis, measuring instructions have been developed that enable the quantitative analysis of unknown saturated samples. However, average molecular mass of the mixture must be known.

---

Entwicklung einer quantitativen FI-MS-Methode zur Charakterisierung von gesättigten hoch- und nichtsiedenden Kohlenwasserstoffgemischen

     

Mehrstufiges Verfahren zur Herstellung von Deuteropropanon auf katalytischem Weg

A novel equipment for a multistep deuteration process of organic compounds was developed. As an example the deuteration of propanone by deuterium oxide over magnesia and thoria was examined. The degree of deuteration of the organic compound and of the catalysts was determined by on-line mass-spectrometry. Percentages of deuterium uptake of the organic compound up to 90% were found. These degrees of deuteration were higher than those measured at the catalysts' surface. This fact can be explained by the kinetic isotope effect.