Stochastic volatility models with leverage and heavy-tailed distributions: A Bayesian approach using scale mixtures

This paper studies a heavy-tailed stochastic volatility (SV) model with leverage effect, where a bivariate Student-t distribution is used to model the error innovations of the return and volatility equations. Choy et al. (2008) studied this model by expressing the bivariate Student-t distribution as a scale mixture of bivariate normal distributions. We propose an alternative formulation by first deriving a conditional Student-t distribution for the return and a marginal Student-t distribution for the log-volatility and then express these two Student-t distributions as a scale mixture of normal (SMN) distributions. Our approach separates the sources of outliers and allows for distinguishing between outliers generated by the return process or by the volatility process, and hence is an improvement over the approach of Choy et al. (2008). In addition, it allows an efficient model implementation using the WinBUGS software. A simulation study is conducted to assess the performance of the proposed approach and its comparison with the approach by Choy et al. (2008). In the empirical study, daily exchange rate returns of the Australian dollar to various currencies and daily stock market index returns of various international stock markets are analysed. Model comparison relies on the Deviance Information Criterion and convergence diagnostic is monitored by Geweke’s convergence test.     

On Constructing Unit Triangular Matrices with Prescribed Singular Values

We propose an efficient algorithm for computing a unit lower triangular n×n matrix with prescribed singular values of O(n ²) cost. This is a solution of the question raised by N. J. Higham in [4, Problem 26.3, p. 528]. Received July 21, 1999; revised November 4, 1999     

Optimal Average Case Estimation in Hilbert Norms

In contrast to the worst case approach, the average case setting provides less conservative insight into the quality of estimation algorithms. In this paper we consider two local average case error measures of algorithms based on noisy information, in Hilbert norms in the problem element and information spaces. We define the optimal algorithm and provide formulas for its two local errors, which explicitly exhibit the influence of factors such as information, information (measurement) errors, norms in the considered spaces, a subset where approximations are allowed, and “unmodeled dynamics.” Based on the error expression, we formulate in algebraic language the problem of selecting the optimal approximating subspace. The solution is given along with the specific formula for the error, which depends on the eigenvalues of a certain matrix defined by information and norms under consideration. Date received: November 25, 1999. Date revised: May 30, 2000.     

Human factors in quality management

This article shows the results of research regarding the importance and the role of human factors in quality management in production enterprises. In creating the concept of systematizations of human factors in quality management an anthropocentrism rule was accepted and resulting from it a category of life quality. Acceptance of this rule led to a multiaspect and comprehensive approach to human factors in quality management. An interpretation of the notion of quality was made, adequate to needs, goals, and requirements of a human. Additionally, a cybernetic model of quality management system was elaborated. Human factors was exhibited in the manager's and executor's roles in quality management systems in enterprise as well as to the roles of creators, producers, and users (customers) in product life cycles. An existence of human factors in ISO norms regarding quality management, work safety, and hygiene management and environment management was also shown. The objective of this article is a comprehensive identification and systematization of aspects of human factors in quality management. © 2008 Wiley Periodicals, Inc.     

Identification of Human Postural Sway

The output of the human postural control system is studied by means of linear system theory. It is assumed that the act of maintaining an erect posture can be treated as an autoregressive system with white noise on input. The identification is performed on the basis of the centre of pressure trajectory and it is shown that the most important features of the postural signal are sufficiently well reproduced by a low order linear autoregressive model. It is shown that the parameters of a model depend on the human subject and in some way characterize his state. Poles of the transmitancy function lying close to the unit circle are discussed as parameters describing the response function and human reactions to external perturbations. In additionf, an analysis of the correlation function within the autoregressive model is performed and its scaling exponents are computed and compared to the experimental results.     

Sterol and steryl ester regulation of phospholipase A2 from the mosquito parasiteLagenidium giganteum

Lagenidium giganteum, a facultative parasite of mosquito larvae, cannot synthesize sterols, and requires an exogenous source of these lipids in order to enter its reproductive cycle. This parasite grows vegetatively in the absence of sterols, but requires cholesterol or structurally related compounds to produce motile zoospores, which are the only stage capable of infecting mosquitoes. Sterols structurally related to cholesterol and some steryl esters inhibited the activity ofL. giganteum phospholipase A₂ (PLA₂), an enzyme that hydrolyzes fatty acids from thesn-2 position of glycerophospholipids. Sterols that induce reproduction inhibitedL. giganteum PLA₂ activity, while sterols and steroids that do not support sporulation had minimal effect. Most steryl esters had no effect on enzyme activity, but cholesteryl arachidonate (CA) was a potent inhibitor of parasite PLA₂. Not all enzymes partly purified using a DEAE-Sephacel column were affected by these lipids, demonstrating selective inhibition of specific enzymes. Potency was enhanced by up to several orders of magnitude if epoxy fatty acids were esterified to the cholesterol nucleus. The steryl ester pool was dynamic during morphogenesis, and the fatty acid composition of the steryl esters did not mimic total cell or membrane (glycerophospholipid) fatty acid composition asL. giganteum proceeded through its growth cycle. Synthesis of CA and monoepoxy CA by the parasite was confirmed using electrospray mass spectrometry and collision-induced dissociation. Steryl derivatives selectively inhibited PLA₂ enzymes from bovine pancreas, snake venom, and human cytoplasmic 85-kDa PLA₂.