Estimating a reduced rank regression model for non-normal variables

A method using third order moments for estimating the regression coefficients as well as the latent state scores of the reduced-rank regression model when the latent variable(s) are non-normally distributed is presented in this paper. It is shown that the factor analysis type indeterminacy of the regression coefficient matrices is eliminated. A real life example of the proposed method is presented. Differences of this solution with the reduced-rank regression eigen solution are discussed.     

Stationary turbulent closure via the Hopf functional equation

Exact closed-form solutions are exhibited for the Hopf equation for stationary incompressible 3D Navier-Stokes flow, for the cases of homogeneous forced flow (including a solution with depleted nonlinearity) and inhomogeneous flow with arbitrary boundary conditions. This provides an exact method for computing two- and higher-point moments, given the mean flow.     

A Recursive Algorithm for Order Cycle-time that Minimizes Logistics Cost

A firm's logistics cost, including shipping and inventory-carrying, is a substantial percentage of its sales. Nevertheless, typical inventory-control methods ignore or insufficiently represent the shipping cost. This paper describes a recursive algorithm that determines the reorder cycle-time that minimizes total logistics cost. It allows for a realistic accounting of shipping cost, which is modelled here as a function of shipping distance and weight. The algorithm uses a relaxation procedure to identify a suitable initial approximation to the optimal order cycle-time and then, through a series of recursive steps, moves to the optimal result. We demonstrate the algorithm with a single item, with a group of items that share a common order cycle, and with multi-items when item demands are random variables. Experience with this algorithm indicates that it converges to the optimal result in a very few steps.     

Study of the reaction238U (α, α′f) atE α=480 MeV

The excitation function of the fission probability P _E E _x) for²³⁸U has been measured in the reaction²³⁸U(α, α′ f) at 480 MeV bombarding energy. The reaction mechanism of this reaction is discussed for excitation energies belowB _nf , the threshold for second chance fission, and aboveB _nf up toE _x =37 MeV. In comparing with results from fission induced by photons and by particle transfer reactions the (α, α′f) reaction gives too low values for the fission probabilityP _f at excitation energies well aboveB _nE . The role of the quasi-elastic knock-out process in this reaction is discussed.     

τ-CHEBYSHEV AND τ-COCHEBYSHEV SUBPSACES OF BANACH SPACES

The concepts of quasi-Chebyshev and weakly-Chebyshev and σ-Chebyshev were defined [3 - 7], andas a counterpart to best approximation in normed linear spaces, best coapprozimation was introduced by Franchetti and Furi^[1]. In this research, we shall define τ-Chebyshev subspaces and τ-cochebyshev subspaces of a Banach space, in which the property τ is compact or weakly-compact, respectively. A set of necessary and sufficient theorems under which a subspace is τ-Chebyshev is defined.     

Schmidt analysis of pure-state entanglement

We examine the application of Schmidt mode analysis to pure-state entanglement. Several examples permitting exact analytic calculation of Schmidt eigenvalues and eigenfunctions are included, as well as evaluation of the associated degree of entanglement.     

The square lattice shuffle

We show that the operations of permuting columns and rows separately and independently mix a square matrix in constant time. © 2006 Wiley Periodicals, Inc. Random Struct. Alg., 2006     

Asymptotic optimality of the bayes estimator in models differentiable in quadratic mean

We study the asymptotic properties of the Bayes estimator in models differentiable in quadratic mean (DQM) in the case of independent and identically distributed observations. The goal is to define weak assumptions on the model under which this estimator is asymptotically efficient, regular, and asymptotically of minimal risk. The results of the paper are applied to models based on a mixture distribution, the Cauchy distribution with location and scale parameter, and the Weibull distribution. Bibliography: 10 titles. Published in Zapiski Nauchnykh Seminarov POMI, Vol. 328, 2005, pp. 114–146.     

A study of structural, compositional and optical properties of spray-deposited non-stoichiometric (Zn,Fe)S thin films

Non-stoichiometric ternary chalcogenides (Zn,Fe)S were prepared in the film form by pyrolytic spray deposition technique, using air/nitrogen as the carrier gas. The precursor solution comprised of ZnCl₂, FeCl₂ and thiourea. The depositions were carried out under optimum conditions of experimental parameters viz. carrier gas (air/nitrogen) flow rate, concentration of precursor constituents, nozzle substrate distance and temperature of quartz substrate. The deposited thin films were later sintered in argon at 1073 K for 120 min.The structural, compositional and optical properties of the sintered thin films were studied. X-ray diffraction studies of the thin films indicated the presence of (Zn,Fe)S solid solution with prominent cubic sphalerite phase while surface morphology as determined by scanning electron microscopy (SEM) revealed a granular structure.The chemical composition of the resulting thin films as analyzed by energy dispersive X-ray analysis (EDAX) reflected the composition of the precursor solutions from which the depositions were carried out with Fe at% values ranging from 0.4 up to 33.SEM micrographs of thin films reveal that the grain sizes of the thin films prepared using air as carrier gas and N₂ as carrier gas are in the vicinity of 300 and 150 nm, respectively.The diffuse transmittance measurements for thin films, as a function of wavelength reveal the dependence of direct optical band gap on Fe content and type of phase.     

Study of thermal currents in powder β-spodumene ceramics doped with Cuo, FeO and TiO2

Ternary-phase ceramic system of Li₂O Al₂O₃ 4SiO₂ doped with CuO, FeO and TiO₂ has been prepared and subjected to dc electrical conductivity and thermally stimulated depolarization current (TSDC) measurements as a function of temperature (30-250 °C) and field strength. The electrical conductivity results are explained by assuming both ionic and electronic conduction mechanisms coexist with different contributions over the whole temperature range of experiments. TSDC spectra have been found to be characterized by a broad intense relaxation peak, which can be attributed to an ionic charge polarization. The broad relaxation transitions are apparently a result of the nonuniform nature of this process. Activation energies are calculated for both dc electrical conductivity and TSDC according to Arrhenius equation and initial rise method, respectively.