Ernst Stadlober

Combined body plethysmographic, spirometric and flow volume reference values for male and female children aged 6 to 16 years obtained from "hospital normals".

Abstract Vital capacity (VC) and its subdivisions (IC and ERV), total lung capacity (TLC), residual volume (RV), peak expiratory flow (PEF), forced vital capacity (FVC), forced expiratory volume in one second (FEV1), maximum flow volume curve (MEF75, MEF50, MEF25, MMEF, FEF75–85), airway resistance (Rtot, Reff) and the thoracic gas volume at resting expiratory position (FRC) were measured in 187 girls and 213 boys (hospital normals) aged 6 to 16 years. The measurements were carried out consecutively on the same subjects in the morning using a volume-constant plethysmograph (MasterLab, E. Jaeger; programs: body plethysmography, spirometry and flow volume). Using multiple regression analysis the best fitting curves for the prediction of normal values for boys and girls were selected. Analyses of covariance were performed to compare the adjusted means of the spirometric and body plethysmographic variables of the male and female subjects. As expected, we found higher static and dynamic (FVC, FEV1, PEF) lung volumes in boys than in girls relating to height. The flows (MMEF, MEF50, MEF25, FEF75–85) were significantly lower in the male than in the female subjects of the same age justifying separate prediction equations, but the same equation for both genders may be used for the resistance variables Rtot and Reff. Our results are compared with those of previous studies. Conclusion Lung volumes and flows differ significantly between girls and boys calling for separate reference values for female and male subjects of the same age.

The ratio of uniforms approach for generating discrete random variates

Abstract The most efficient algorithms for sampling from classical discrete distributions are based upon the acceptance/rejection principle. They are complicated and not easy to understand. By adapting the ratio of uniforms method to unimodal discrete distributions, sampling procedures can be established, which are both simple and fast. Algorithms for the hypergeometric distribution are developed and compared with competing methods.

Generating beta variates via patchwork rejection

A new algorithm for sampling from beta(p, q) distributions with parametersp>1,q>1 is developed. It is based on a method by Minh [9] which improves acceptance-rejection sampling in the main part of the distributions. Additionally, transformed uniform deviates can often be accepted immediately, so that much fewer than two uniforms are needed for one beta variate, on the average. The remaining tests for acceptance are enhanced by ‘squeezes’. Experiments covering a wide range of pairs (p, q) showed improvements in speed over competing algorithms in most cases.ZusammenfassungEin neuer Algorithmus zur Erzeugung von Stichproben aus Beta(p, q)-Verteilungen mit Parameternp>1,q>1 wird entwickelt. Er basiert auf einer Methode von Minh [9], welche die Verwerfungstechnik im Hauptteil der Verteilungen verbessert. Zusätzlich können transformierte gleichverteilte Zufallszahlen oft sofort akzeptiert werden, so daß im Durchschnitt viel weniger als zwei gleichverteilte Zufallszahlen pro Beta-Stichprobe benötigt werden. Die verbleibenden Annahmetests werden durch “Squeezes” verbessert. Versuche über einen weiten (p, q)-Bereich ergaben in den meisten Fällen Geschwindigkeitssteigerungen im Vergleich zu konkurrierenden Methoden.

Quantitative Text Typology: The Impact of Word Length

The present study aims at the quantitative classification of texts and text types. By way of a case study, 398 Slovenian texts from different genres and authors are analyzed as to their word length. It is shown that word length is an important factor in the synergetic self-regulation of texts and text types, and that word length may significantly contribute to a new typology of discourse types.2

Classification of Author and/or Genre? The Impact of Word Length

190 Russian texts — letters and poems by three different authors — are analyzed as to their word length. The basic question concerns the quantitative classification of these texts as to authorship or as to text sort. By way of multivariate analyses it is shown that word length is a characteristic of genre, rather than of authorship.3

SPICE modeling of process variation using location depth corner models

For robust designs, the influence of process variations has to be considered during circuit simulation. We propose a nonparametric statistical method to find sets of simulation parameters that cover the process spread with a minimum number of simulation runs. Process corners are determined from e-test parameter vectors using a location depth algorithm. The e-test corner vectors are then transformed to SPICE parameter vectors by a linear mapping. A special corner extension algorithm makes the resulting simulation setup robust against moderate process shifts while preserving the underlying correlation structure. To be applicable in a production and circuit design environment, the models are integrated into an automated model generation flow for usage within a design-framework. The statistical methods are validated for analog/mixed-signal benchmark circuits.

The patchwork rejection technique for sampling from unimodal distributions

We report on both theoretical developments and comutational experience with the patchwork rejection technique in Zechner and Stadlober [1993] and Zechner [1997]. The basic approach is due to Minh [1988], who suggested a special sampling method for the gamma distribution. The methods general objective is to rearrange the area below the density of histogram f (x) in the body of the distribution by certain point reflections such that variates may be generated efficiently within a large center interval. This is carried out via uniform hat functions, combined with minorizing rectangles for immediate acceptance of one transformed uniform deviate. The remaining tails of f(x) are covered by exponential functions. Experiments show that patchwork rejection algorithms are in general faster than their competitors at the cost of higher set-up times.

METHODS FOR ESTIMATING PRINCIPAL POINTS

ABSTRACT Principal points of a distribution have been introduced by Flury [1] who tackled the problem of optimal grouping in multivariate data. In essence, principal points are the theoretical counterparts of cluster means obtained by a k-means clustering algorithm. There has been considerable effort to find efficient estimation procedures for principal points. It is well known that under certain conditions the k-means estimator is a consistent and asymptotically normal estimator of the population principal points. In this paper some material on principal points is reviewed and new algorithms for the estimation of principal points in univariate distributions (univariate principal points) are proposed. Additionally, the Bootstrap approach is applied to assess the variability of the suggested estimators.

Quantitative Text Typology: The Impact of Sentence Length

This study focuses on the contribution of sentence length for a quantitative text typology. Therefore, 333 Slovenian texts are analyzed with regard to their sentence length. By way of multivariate discriminant analyses (M D A) it is shown that indeed, a text typology is possible, based on sentence length, only; this typology, however, does not coincide with traditional text classifications, such as, e.g., text sorts or functional style. Rather, a new categorization into specific discourse types seems reaonable.

The Relationship of Word Length and Sentence Length: The Inter-Textual Perspective

The present study concentrates on the relation between sentence length (SL) and word length (WL) as a possible factor in text classification. The dependence of WL and SL is discussed in terms of general system theory and synergetics; the results achieved thus are relevant not only for linguistic studies of text classification, but for the study of other complex systems, as well.