Multi-sample analysis of moment-structures: Asymptotic validity of inferences based on second-order moments

In moment structure analysis with nonnormal data, asymptotic valid inferences require the computation of a consistent (under general distributional assumptions) estimate of the matrix $\Gamma$ of asymptotic variances of sample second--order moments. Such a consistent estimate involves the fourth--order sample moments of the data. In practice, the use of fourth--order moments leads to computational burden and lack of robustness against small samples. In this paper we show that, under certain assumptions, correct asymptotic inferences can be attained when $\Gamma$ is replaced by a matrix $\Omega$ that involves only the second--order moments of the data. The present paper extends to the context of multi--sample analysis of second--order moment structures, results derived in the context of (simple--sample) covariance structure analysis (Satorra and Bentler, 1990). The results apply to a variety of estimation methods and general type of statistics. An example involving a test of equality of means under covariance restrictions illustrates theoretical aspects of the paper.

Scaled and adjusted restricted tests in multi-sample analysis of moment structures

We extend to score, Wald and difference test statistics the scaled and adjusted corrections to goodness-of-fit test statistics developed in Satorra and Bentler (1988a,b). The theory is framed in the general context of multisample analysis of moment structures, under general conditions on the distribution of observable variables. Computational issues, as well as the relation of the scaled and corrected statistics to the asymptotic robust ones, is discussed. A Monte Carlo study illustrates thecomparative performance in finite samples of corrected score test statistics.

Bringing game theory to hypothesis testing: Establishing finite sample bounds on inference

Small sample properties are of fundamental interest when only limited data is avail-able. Exact inference is limited by constraints imposed by speci.c nonrandomizedtests and of course also by lack of more data. These e¤ects can be separated as we propose to evaluate a test by comparing its type II error to the minimal type II error among all tests for the given sample. Game theory is used to establish this minimal type II error, the associated randomized test is characterized as part of a Nash equilibrium of a .ctitious game against nature.We use this method to investigate sequential tests for the di¤erence between twomeans when outcomes are constrained to belong to a given bounded set. Tests ofinequality and of noninferiority are included. We .nd that inference in terms oftype II error based on a balanced sample cannot be improved by sequential sampling or even by observing counter factual evidence providing there is a reasonable gap between the hypotheses.

Some hypothesis tests for the covariance matrix when the dimension is large compared to the sample size

This paper analyzes whether standard covariance matrix tests work whendimensionality is large, and in particular larger than sample size. Inthe latter case, the singularity of the sample covariance matrix makeslikelihood ratio tests degenerate, but other tests based on quadraticforms of sample covariance matrix eigenvalues remain well-defined. Westudy the consistency property and limiting distribution of these testsas dimensionality and sample size go to infinity together, with theirratio converging to a finite non-zero limit. We find that the existingtest for sphericity is robust against high dimensionality, but not thetest for equality of the covariance matrix to a given matrix. For thelatter test, we develop a new correction to the existing test statisticthat makes it robust against high dimensionality.

Honey, I shrunk the sample covariance matrix

The central message of this paper is that nobody should be using the samplecovariance matrix for the purpose of portfolio optimization. It containsestimation error of the kind most likely to perturb a mean-varianceoptimizer. In its place, we suggest using the matrix obtained from thesample covariance matrix through a transformation called shrinkage. Thistends to pull the most extreme coefficients towards more central values,thereby systematically reducing estimation error where it matters most.Statistically, the challenge is to know the optimal shrinkage intensity,and we give the formula for that. Without changing any other step in theportfolio optimization process, we show on actual stock market data thatshrinkage reduces tracking error relative to a benchmark index, andsubstantially increases the realized information ratio of the activeportfolio manager.

Nonlinear models and small sample performance of the generalized method of moments

In this paper I explore the issue of nonlinearity (both in the datageneration process and in the functional form that establishes therelationship between the parameters and the data) regarding the poorperformance of the Generalized Method of Moments (GMM) in small samples.To this purpose I build a sequence of models starting with a simple linearmodel and enlarging it progressively until I approximate a standard (nonlinear)neoclassical growth model. I then use simulation techniques to find the smallsample distribution of the GMM estimators in each of the models.

An inequality for uniform deviations of sample averages from their means

We derive a new inequality for uniform deviations of averages from their means. The inequality is a common generalization of previous results of Vapnik and Chervonenkis (1974) and Pollard (1986). Usingthe new inequality we obtain tight bounds for empirical loss minimization learning.

Gas collisions and pressure quenching of the photoluminescence of silicon nanopowder grown by plasma-enhanced chemical vapor deposition

The quenching of the photoluminescence of Si nanopowder grown by plasma-enhanced chemical vapor deposition due to pressure was measured for various gases ( H2, O2, N2, He, Ne, Ar, and Kr) and at different temperatures. The characteristic pressure, P0, of the general dependence I(P) = I0¿exp(¿P/P0) is gas and temperature dependent. However, when the number of gas collisions is taken as the variable instead of pressure, then the quenching is the same within a gas family (mono- or diatomic) and it is temperature independent. So it is concluded that the effect depends on the number of gas collisions irrespective of the nature of the gas or its temperature.

Linear and nonlinear optical properties of Si nanocrystals in SiO2 deposited by plasma-enhanced chemical-vapor deposition

Linear and nonlinear optical properties of silicon suboxide SiOx films deposited by plasma-enhanced chemical-vapor deposition have been studied for different Si excesses up to 24¿at.¿%. The layers have been fully characterized with respect to their atomic composition and the structure of the Si precipitates. Linear refractive index and extinction coefficient have been determined in the whole visible range, enabling to estimate the optical bandgap as a function of the Si nanocrystal size. Nonlinear optical properties have been evaluated by the z-scan technique for two different excitations: at 0.80¿eV in the nanosecond regime and at 1.50¿eV in the femtosecond regime. Under nanosecond excitation conditions, the nonlinear process is ruled by thermal effects, showing large values of both nonlinear refractive index (n2 ~ ¿10¿8¿cm2/W) and nonlinear absorption coefficient (ß ~ 10¿6¿cm/W). Under femtosecond excitation conditions, a smaller nonlinear refractive index is found (n2 ~ 10¿12¿cm2/W), typical of nonlinearities arising from electronic response. The contribution per nanocrystal to the electronic third-order nonlinear susceptibility increases as the size of the Si nanoparticles is reduced, due to the appearance of electronic transitions between discrete levels induced by quantum confinement.

Properties of nitrogen doped silicon films deposited by low-pressure chemical vapor deposition from silane and ammonia

Nitrogen doped silicon (NIDOS) films have been deposited by low-pressure chemical vapor deposition from silane SiH4 and ammonia NH3 at high temperature (750°C) and the influences of the NH3/SiH4 gas ratio on the films deposition rate, refractive index, stoichiometry, microstructure, electrical conductivity, and thermomechanical stress are studied. The chemical species derived from silylene SiH2 into the gaseous phase are shown to be responsible for the deposition of NIDOS and/or (silicon rich) silicon nitride. The competition between these two deposition phenomena leads finally to very high deposition rates (100 nm/min) for low NH3/SiH4 gas ratio (R¿0.1). Moreover, complex variations of NIDOS film properties are evidenced and related to the dual behavior of the nitrogen atom into silicon, either n-type substitutional impurity or insulative intersticial impurity, according to the Si¿N atomic bound. Finally, the use of NIDOS deposition for the realization of microelectromechanical systems is investigated.

Microstructure of highly oriented, hexagonal, boron nitride thin films grown on crystalline silicon by radio frequency plasma-assisted chemical vapor deposition

We present a high‐resolution electron microscopy study of the microstructure of boron nitride thin films grown on silicon (100) by radio‐frequency plasma‐assisted chemical vapor deposition using B2H6 (1% in H2) and NH3 gases. Well‐adhered boron nitride films grown on the grounded electrode show a highly oriented hexagonal structure with the c‐axis parallel to the substrate surface throughout the film, without any interfacial amorphous layer. We ascribed this textured growth to an etching effect of atomic hydrogen present in the gas discharge. In contrast, films grown on the powered electrode, with compressive stress induced by ion bombardment, show a multilayered structure as observed by other authors, composed of an amorphous layer, a hexagonal layer with the c‐axis parallel to the substrate surface and another layer oriented at random

Thin silicon films ranging from amorphous to nanocrystalline obtained by Hot-Wire CVD

In this paper, we have presented results on silicon thin films deposited by hot-wire CVD at low substrate temperatures (200 °C). Films ranging from amorphous to nanocrystalline were obtained by varying the filament temperature from 1500 to 1800 °C. A crystalline fraction of 50% was obtained for the sample deposited at 1700 °C. The results obtained seemed to indicate that atomic hydrogen plays a leading role in the obtaining of nanocrystalline silicon. The optoelectronic properties of the amorphous material obtained in these conditions are slightly poorer than the ones observed in device-grade films grown by plasma-enhanced CVD due to a higher hydrogen incorporation (13%).

Gas collisions and pressure quenching of the photoluminescence of silicon nanopowder grown by plasma-enhanced chemical vapor deposition

The quenching of the photoluminescence of Si nanopowder grown by plasma-enhanced chemical vapor deposition due to pressure was measured for various gases ( H2, O2, N2, He, Ne, Ar, and Kr) and at different temperatures. The characteristic pressure, P0, of the general dependence I(P) = I0¿exp(¿P/P0) is gas and temperature dependent. However, when the number of gas collisions is taken as the variable instead of pressure, then the quenching is the same within a gas family (mono- or diatomic) and it is temperature independent. So it is concluded that the effect depends on the number of gas collisions irrespective of the nature of the gas or its temperature.

Plasma-enhanced chemical vapor deposition of boron nitride thin films from B2H6-H2-NH3 and B2H6-N2 gas mixtures

Highly transparent and stoichiometric boron nitride (BN) films were deposited on both electrodes (anode and cathode) of a radio-frequency parallel-plate plasma reactor by the glow discharge decomposition of two gas mixtures: B2H6-H2-NH3 and B2H6-N2. The chemical, optical, and structural properties of the films, as well as their stability under long exposition to humid atmosphere, were analyzed by x-ray photoelectron, infrared, and Raman spectroscopies; scanning and transmission electron microscopies; and optical transmittance spectrophotometry. It was found that the BN films grown on the anode using the B2H6-H2-NH3 mixture were smooth, dense, adhered well to substrates, and had a textured hexagonal structure with the basal planes perpendicular to the film surface. These films were chemically stable to moisture, even after an exposition period of two years. In contrast, the films grown on the anode from the B2H6-N2 mixture showed tensile stress failure and were very unstable in the presence of moisture. However, the films grown on the cathode from B2H6-H2-NH3 gases suffered from compressive stress failure on exposure to air; whereas with B2H6-N2 gases, adherent and stable cathodic BN films were obtained with the same crystallographic texture as anodic films prepared from the B2H6-H2-NH3 mixture. These results are discussed in terms of the origin of film stress, the effects of ion bombardment on the growing films, and the surface chemical effects of hydrogen atoms present in the gas discharge.