A semiparametric model for compositional data analysis in presence of covariates on the simplex

Compositional data occur as natural realizations of multivariate observations comprising element proportions of some whole quantity. Such observations predominate in disciplines like geology, biology, ecology, economics and chemistry. Due to unit sum constraint on compositional data, specialized statistical methods are required for analyzing these data. Dirichlet distributions were originally used to study compositional data even though this family of distribution is not appropriate (see Aitchison, 1986) because of their extreme independence properties. Aitchison (1982) endeavored to provide a viable alternative to existing methods by employing Logistic Normal distribution to analyze such constrained data. However this family does not include the Dirichlet class and is therefore unable to address the issue of extreme independence. In this paper generalized Liouville family is investigated to model compositional data which includes covariates. This class permits distributions that admit negative or mixed correlation and also contains non-Dirichlet distributions with non-positive correlation and overcomes deficits in the Dirichlet class. Semiparametric Bayesian methods are proposed to estimate the probability density. Predictive distributions are used to assess performance of the model. The methods are illustrated on a real data set.     

Bending and free vibration analysis of shear deformable laminated composite beams by finite element method

In the present investigation a higher-order shear deformation theory and the conventional first-order theory are used to develop a finite element method to analyse accurately the bending and free vibration behaviour of laminated composite beams, using nine-noded isoparametric elements. The higher-order theory assumes all the displacement components, u, v and w, which contain variation up to a cubic power of z. The effects of various parameters such as fibre orientation, stacking sequence, span-to-thickness ratio and support condition on the non-dimensionalised deflections, stresses and fundamental frequencies are investigated. Cases where only the higher-order theory is likely to yield accurate results are highlighted.     

Corrosion failure of Zn–Al detonator housing

Some batches of detonator housings made up of Chromium plated Zn–Al alloy were found in an extensively cracked condition after few months of storage at room temperature. An analysis of the failure showed that the cracks were due to intergranular corrosion facilitated by segregation of lead at the grain boundaries. Improper chromium plating further aggravated the corrosion problem. This failure case emphasises the need for strict control of chemical composition for components made from Zn–Al alloy and the process of Cr plating of the components.     

Engineering a lever into the kinesin neck

To probe for a lever arm action in the kinesin stepping mechanism, we engineered a rodlike extension piece into the tail of rat kinesin at various points close to the head-tail junction and measured its effects on the temperature dependence of velocity in microtubule gliding assays. The insert comprised two contiguous alpha-actinin triple-coil repeats and was predicted to fold into a stiff rodlike module about 11 nm long. The effects of this module were greater the closer it was placed to the head-tail junction. When inserted distal to the head-tail junction, at Asn401 in the dimeric K partial differential401GST, the insert had no effect. When inserted closer to the heads at Val376 into K partial differential376GST, the insert slowed progress below 22 degreesC but accelerated progress to approximately 125% of wild type above 22 degreesC. The most dramatic effect of the synthetic lever occurred when it was inserted very close to the head-neck junction, at Glu340 into the single-headed construct K partial differential340GST. This construct was immotile without the insert, but motile with it, at about 30% of the velocity of the dimeric control. The alpha-actinin module thus confers some gain-of-function when inserted close to the head-neck junction but not when placed distal to it. The data exclude the presence of a lever arm C-terminal to Val376 in the kinesin tail but suggest that a short-throw lever arm may be present, N-terminal to Val376 and contiguous with the head-neck junction at Ala339.     

GPU accelerated novel particle filtering method

In this paper, a graphics processor unit (GPU) accelerated particle filtering algorithm is presented with an introduction to a novel resampling technique. The aim remains in the mitigation of particle impoverishment as well as computational burden, problems which are commonly associated with classical (systematic) resampled particle filtering. The proposed algorithm employs a priori-space dependent distribution in addition to the likelihood, and hence is christened as dual distribution dependent (D3) resampling method. Simulation results exhibit lesser values for root mean square error (RMSE) in comparison to that for systematic resampling. D3 resampling is shown to improve particle diversity after each iteration, thereby affecting the overall quality of estimation. However, computational burden is significantly increased owing to few excessive computations within the newly formulated resampling framework. With a view to obtaining parallel speedup we introduce a CUDA version of the proposed method for necessary acceleration by GPU. The GPU programming model is detailed in the context of this paper. Implementation issues are discussed along with illustration of empirical computational efficiency, as obtained by executing the CUDA code on Quadro 2000 GPU. The GPU enabled code has a speedup of 3 and 4 over the sequential executions of systematic and D3 resampling methods respectively. Performance both in terms of RMSE and running time have been elaborated with respect to different selections for threads per block towards effective implementations. It is in this context that, we further introduce a cost to performance metric (CPM) for assessing the algorithmic efficiency of the estimator, involving both quality of estimation and running time as comparative factors, transformed into a unified parameter for assessment. CPM values for estimators obtained from all such different choices for threads per block have been determined and a final value for the chosen parameter is resolved for generation of a holistic effective estimator.     

Grafting of methyl methacrylate onto guar gum by hydrogen peroxide initiation

Graft copolymerization of methyl methacrylate (MMA) onto guar gum (GG) in aqueous slurry has been carried out using hydrogen peroxide (H₂O₂) as initiator. The copolymers were characterized by infrared spectroscopy. The grafting parameters like percent grafting, grafting efficiency, percent add-on, and the grafting frequency were determined, and the effect of reaction time, concentration of initiator, and [GG]/[MMA] ratios on the grafting parameters have been discussed. The decrease in % add-on at increasing concentration of H₂O₂ indicated enhancement in the rate of homopolymerization of methyl methacrylate.     

Natural rubber–ethylene‐propylene‐diene rubber covulcanization: Effect of reinforcing fillers

Filled covulcanizates of elastomer blend comprising natural rubber (NR) and ethylene‐propylene‐diene rubber (EPDM) of commercial importance were successfully prepared by using a multifunctional rubber additive; namely, bis(diisopropyl)thiophosphoryl disulfide (DIPDIS). A Two‐stage vulcanization technique further improved the physicochemical properties of the blend vulcanizates by restricting, through the formation of polar rubber bound intermediates, the migration of curative and filler from lower to highly unsaturated rubber. Scanning electron microscopy studies indicate homogeneity and coherency in the morphology of the two‐stage vulcanizates. © 2002 Wiley Periodicals, Inc. J Appl Polym Sci 84: 1001–1010, 2002; DOI 10.1002/app.10361     

Effect of thiophosphoryl compounds as accelerators and coupling agents on the vulcanization of NR‐XNBR blend

The effect of a number of thiophosphoryl compounds [viz., bis(isobutyl) thiophosphoryl di‐, tri‐, and tetrasulfides and bis(dicyclohexyl) thiophopsphoryl di‐, tri‐, and tetrasulfides] on natural rubber (NR)‐carboxylated nitrile rubber (XNBR) blend was studied. All these thiophosphoryl compounds are capable of forming interrubber links leading to covulcanized blends which exhibit a fair degree of synergism with respect to physical properties, the maximum being obtained at the 25% NR and 75% XNBR blend composition. This is an obvious claim that the blend under investigation is technologically compatible, having some degree of interrubber interaction which is enhanced in case of two‐stage vulcanization. The existence of interrubber interaction is judged by the swelling experiment. The blend morphology assessed by SEM micrographs corroborates the foregoing observations and accounts for the significant improvement in physical properties of the blend vulcanizates, particularly in two‐stage vulcanization. © 2002 Wiley Periodicals, Inc. J Appl Polym Sci 86: 3286‐3299, 2002