Power and Measures of Effect Size in Analysis of Variance With Fixed Versus Random Nested Factors.

Ignoring a nested factor can influence the validity of statistical decisions about treatment effectiveness. Previous discussions have centered on consequences of ignoring nested factors versus treating them as random factors on Type I errors and measures of effect size (B. E. Wampold & R. C. Serlin, see record 2000-16737-003). The authors (a) discuss circumstances under which the treatment of nested provider effects as fixed as opposed to random is appropriate; (b) present 2 formulas for the correct estimation of effect sizes when nested factors are fixed; (c) present the results of Monte Carlo simulations of the consequences of treating providers as fixed versus random on effect size estimates, Type I error rates, and power; and (d) discuss implications of mistaken considerations of provider effects for the study of differential treatment effects in psychotherapy research. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Phase equilibria calculations for ill-defined mixtures using the prsv equation of state

A method for the calculation of vapor-liquid equilibria (VLE) for multicomponent systems using continuous thermodynamics is presented. The Extended Spline Fit Technique (ESFT) is used to represent the molar distribution curve of petroleum fluids. A continuous version of the Peng-Robinson-Stryjek-Vera (PRSV) equation of state (EOS) with Generalized Single Carbon Number Properties (GSCNP) is utilized. For continuously distributed components, following the suggestion of Hendriks (1987), the number of equations solved is reduced through integration over the range of the distributing variable using Legendre-Gauss quadrature. Integrated equations are solved using accelerated successive substitution. Flash, dew and bubble point calculations for an imaginary and a real system yield satisfactory results.     

Statistical multiplexing of VBR sources: A matrix-analytic approach

In this paper we propose a versatile point process as model for a large class of variable bit rate sources and their superpositions. The process belongs to the class of discrete-time batch Markovian arrival processes (D-BMAP). Its use leads to computationally tractable and accurate solutions for various performance measures of the related statistical multiplexer: buffer occupancy distribution, cell loss probability, conditional cell loss probability. Moreover it is shown that the output process of the multiplexer belongs to the same class of processes.     

An Integrated CAD Methodology for Yield Enhancement of VLSI CMOS Circuits Including Statistical Device Variations

In this paper a novel CAD methodology for yield enhancement of VLSI CMOS circuits including random device variations is presented. The methodology is based on a preliminary characterization of the technological process by means of specific test chips for accurate mismatch modeling. To this purpose, a very accurate position-dependent parameter mismatch model has been formulated and extracted. Finally a CAD tool implementing this model has been developed. The tool is fully integrated in an environment of existing commercial tools and it has been experimented in the STMicroelectronics Flash Memory CAD Group.As an example of application, a bandgap reference circuit has been considered and the results obtained from simulations have been compared with experimental data. Furthermore, the methodology has been applied to the read path of a complex Flash Memory produced by STMicroelectronics, consisting of about 16,000 MOSFETs. Measurements of electrical performances have confirmed the validity of the methodology, and the accuracy of both the mismatch model and the simulation flow.     

On the Convergence of Fitting Algorithms in Computer Vision

We investigate several numerical schemes for estimating parameters in computer vision problems: HEIV, FNS, renormalization method, and others. We prove mathematically that these algorithms converge rapidly, provided the noise is small. In fact, in just 1-2 iterations they achieve maximum possible statistical accuracy. Our results are supported by a numerical experiment. We also discuss the performance of these algorithms when the noise increases and/or outliers are present. Nikolai Chernov PhD in mathematics from Moscow University in 1984. Researcher in JINR (Dubna, Russia) in 1984–91. Professor of Mathematics at University of Alabama at Birmingham, USA, since 1994.     

Modelling the change in colour of potassium sorbate powder during heating

Colour change of potassium sorbate caused by heating was statistically modelled to examine its potential as a time–temperature process indicator. A full factorial design with five levels of heating time ( t ) and temperature ( T ) was used. L , a and b values were measured by a HunterLab colorimeter, and second-order polynomial models were developed. Heating significantly reduced L and increased b values, causing the potassium sorbate powder to turn dark yellow, possibly because of non-enzymatic browning reactions. Statistically significant models with good fit were developed for L and b , and their response surfaces were generated. This potassium sorbate-based system may be a suitable indicator for high-temperature heating processes such as dry heat sterilizers or baking ovens.     

Nonlinear Viscoelastic and Viscoplastic Constitutive Equations Based on Thermodynamics

An approach to modeling the mechanical behavior of fiber reinforced and unreinforced plastics with an evolving internal state is described. Intrinsic nonlinear viscoelastic and viscoplastic behavior of the resin matrix is taken into account along with growth of damage. The thermodynamic framework of the method is discussed first. The Gibbs free energy is expressed in terms of stresses, internal state variables (ISVs), temperatureand moisture content. Simplifications are introduced based on physical models for evolution of the ISVs and on experimental observations of thedependence of strain state on stress state and its history. These simplifications include use of master creep functions that account for multiaxial stresses, environmental factors and aging in a reduced time and other scalars. An explicit representation of the strains follows, which isthen specialized to provide three-dimensional homogenized constitutiveequations for transversely isotropic, fiber composites. Experimentalsupport for these equations is briefly reviewed. Finally, physicalinterpretation of some of the constitutive functions is discussed usingresults from a microcracking model as well as molecular rate process andfree volume theories. It is shown that the present thermodynamicformulation leads to a generalized rate process theory that accounts for abroad distribution of thermally activated transformations in polymers.     

Control charts for processes subject to random shocks

In this paper, a charting technique for controlling processes subject to random shocks is presented. This type of process is common in a high-yield production environment and the conventional Shewhart control charts are not efficient for its monitoring and control. This technique is able to detect process improvement, easy for decision making, and more concise and informative than other methods used for this type of process. In addition, it could provide diagnostic information which is highly useful in practice.     

Is the area measure a historical anomaly?

Green's well-known area theorem establishes an equivalence between the area under the yes-no ROC curve and the percent correct of an unbiased observer in a two-alternative forced-choice (2AFC) task with equivalent stimuli. In this article, we show that this conversion from yes-no detection data to hypothetical performance in a 2AFC task is unnecessary: The same yes-no detection data that are used to compute the area statistic can always be used to compute the percent correct of an unbiased observer in the yes-no detection task itself. We also show that the ROC curve may not be the ideal graphical device for many investigators. A more natural representation of the difficulty of a discrimination task is obtained by plotting the distribution of the posterior betting odds under equal base rates, which can be estimated from their distributions under unequal base rates. Finally, unlike the area measure and other traditional detection theory statistics, both the yes-no percent correct measure and the odds distributions generalize in an obvious and direct way to classification paradigms with more than two responses (e.g., identification). (PsycINFO Database Record (c) 2010 APA, all rights reserved)