Extended redundancy analysis (ERA) is a statistical approach to component-based multivariate regression modeling that explores interrelationships among multiple sets of while incorporating regression with a data-reduction technique. The extant models that utilize ERA have assumed the outcome variables with the same data type. Also, ERA models focused on estimating direct pathways only without explicitly addressing mediation effects. In this paper, ERA is extended to handle multiple mediators and mixed types of outcome variables by adopting a Bayesian framework, taking into account correlation structure among all of the outcome variables. The proposed method develops an algorithm that derives the joint posterior distribution of parameters using a Markov chain Monte Carlo algorithm. Simulations and an empirical dataset are provided to illustrate the usefulness of the proposed method.

Extended redundancy analysis (ERA), a generalized version of redundancy analysis (RA), has been proposed as a useful method for examining interrelationships among multiple sets of variables in multivariate linear regression models. As a limitation of the extant RA or ERA analyses, however, parameters are estimated by aggregating data across all observations even in a case where the study population could consist of several heterogeneous subpopulations. In this paper, we propose a Bayesian mixture extension of ERA to obtain both probabilistic classification of observations into a number of subpopulations and estimation of ERA models within each subpopulation. It specifically estimates the posterior probabilities of observations belonging to different subpopulations, subpopulation-specific residual covariance structures, component weights and regression coefficients in a unified manner. We conduct a simulation study to demonstrate the performance of the proposed method in terms of recovering parameters correctly. We also apply the approach to real data to demonstrate its empirical usefulness.

Dynamic recrystallization (DRX) of 99.9999% aluminum single crystal at room temperature was examined under frictionless deformation mode. To exclude the self-heating of the specimen due to applied high strain, a microcrack that localizes the stress at a very small region was intentionally introduced by controlled local necking. For the in situ observation of DRX, a specially designed in situ microdeformation device was positioned inside an electron backscattered diffraction system chamber. Recrystallized grains showed relatively random texture and preferred growth direction. The subgrains with low-angle grain boundaries formed by dynamic recovery transformed into small grains with high-angle grain boundaries, acting as nuclei for discontinuous dynamic recrystallization and growing by further deformation. The DRX in pure aluminum can take place under frictionless tensile deformation conditions at room temperature, and the stress localization and high purity are key issues for the DRX of aluminum at room temperature.

We consider a single-server retrial queueing system where retrial time is inversely proportional to the number of customers in the system. A necessary and sufficient condition for the stability of the system is found. We obtain the Laplace transform of virtual waiting time and busy period. The transient distribution of the number of customers in the system is also obtained.

The preparation of BaMgAl10O17:Eu2+ (BAM) blue phosphor layer for a plasma display panel by the addition of a newly designed energetic dispersant (hereafter referred to as SAIT7) and its resulting photoluminescence (PL) efficiency is investigated. The addition of SAIT7 increases the maximum solids loading of BAM phosphor paste from 27.91 to 35.87 vol% and yields a highly packed microstructure of the phosphor layer. The PL intensity is increased by 7.57% compared to the BAM blue phosphor layer prepared without SAIT7. In conclusion, the addition of SAIT7 increased the packing density of the phosphor layer and resulted in improved luminescent properties of the phosphor layer.

The present study examines whether illusory movement (IM) of a horizontal line, induced by a moving background (MB), influences line-bisection performance in normal subjects. The first experiment attempted to identify the speeds of MB that induce IM. We found that when speed is increased from 1.53° to 13.3°/sec, IM increases, but that with further speed increases, IM decreases. Leftward MB induces rightward IM, and vice versa. In the second experiment, we had subjects bisect lines at MB speeds that had been shown to induce IM in the first experiment. We found that leftward MB induced a rightward bias, and vice versa. We also found that there was a relationship between the magnitude of IM and the degree of bias. In the third experiment, by making the target line larger than the MB, we made the conditions where IM was presumably absent. Unlike the results of bisection performed with IM, subjects showed a bias in the direction of the MB. Overall, these experiments demonstrated that the perception of motion induces subjects to attend in the direction of movement. (JINS, 2005, 11, 881–888.)

A mucin coat is deposited on rabbit embryos during passage through the oviduct; rabbit blastocysts cultured from the 1-cell stage in vitro have no mucin coat. When cultured blastocysts are transferred to recipients, the lack of mucin coat might account in part for subsequent failure of pregnancy. We have investigated the possibility that mucin coat deposition is induced following transfer of in vitro 72 h-cultured blastocysts to oviducts of asynchronous or synchronous recipients. One-cell embryos were collected by flushing oviducts 19–20 h post-coitus and were cultured in vitro for 72 h until they reached the blastocyst stage. The blastocysts were transferred to the oviducts of recipients that were synchronized either with the donors (synchronous) or 1 day later than the donors (asynchronous). They were recovered after 24–48 h and the mucin coat thickness and embryo degeneration rate were measured. The degeneration rate of blastocysts recovered from uteri of synchronous recipients was higher than that from asynchronous recipients (72.2% vs 40.0%). The mucin coats around embryos recovered from oviducts of asynchronous recipients after 48 h were thicker than those from synchronous recipients. More asynchronous recipients were pregnant and gave birth to more pups than synchronous recipients. These results indicate that the oviducts of asynchronous recipients secreted more mucin around the transferred embryos, causing higher rates of implantation of the in vitro-cultured blastocysts.

Silicon nanocrystals were in situ grown in a silicon nitride film by plasma enhanced chemical vapor deposition. The size and structure of silicon nanocrystals were confirmed by high-resolution transmission electron microscopy. Depending on the size, the photoluminescence of silicon nanocrystals can be tuned from the near infrared (1.38 eV) to the ultraviolet (3.02 eV). The fitted photoluminescence peak energy as E(eV) = 1.16 + 11.8/d2 is an evidence for the quantum confinement effect in silicon nanocrystals. The results demonstrate that the band gap of silicon nanocrystals embedded in silicon nitride matrix was more effectively controlled for a wide range of luminescent wavelengths.

We have studied infrared photoluminescence (PL) and x-ray diffraction (XRD) of 400 nm and 1500 nm thick InAs epilayers on GaAs, and 4 nm thick InAs on graded InGaAs layer with total thickness of 300 nm grown by molecular beam epitaxy. The PL peak positions of 400 nm, 1500 nm and 4 nm InAs epilayer measured at 10 K are blue-shifted from that of InAs bulk by 6.5, 4.5, and 6 meV, respectively, which can be largely explained by the residual strain in the epilayer. The residual strain caused by the lattice mismatch between InAs and GaAs or graded InGaAs/GaAs was observed from XRD measurements. While the PL peak position of 400 nm thick InAs layer is linearly shifted toward higher energy with increase in excitation intensity ranging from 10 to 140 mW, those of 4 nm InAs epilayer on InGaAs and 1500 nm InAs layer on GaAs is gradually blue-shifted and then, saturated above a power of 75 mW. These results suggest that adopting a graded InGaAs layer between InAs and GaAs can efficiently reduce the strain due to lattice mismatch in the structure of InAs/GaAs.

Si delta-doping in the GaN layer has been successfully demonstrated by low-pressure metalorganic chemical vapor deposition at a growth temperature of 1040 . Si delta-doping concentration increases and then decreases with an increase in delta-doping time. This indicates that delta-doping concentration is limited by the desorption process owing to much higher thermal decomposition efficiency of silane at high growth temperatures of GaN. In addition, it was observed that the use of a post-purge step in the ammonia ambient reduces Si delta-doping concentration. From capacitance-voltage measurement, a sharp carrier concentration profile with a full-width at half maximum of 4.1 nm has been achieved with a high peak concentration of 9.8 1018 cm−3.

A stress-driven formation of self-assembled InGaAs islands has been studied by the growth on GaAs (100) substrates with sub-micron platinum stripe pattern. Islands or quantum dots preferentially nucleate at the boundary of metal patterns. In addition, a quantum dot-free region near the boundary of the metal pattern is found. Those results are attributed to the stress between metal stripe and GaAs surface, which produces a laterally stressed region around the metal stripe. Adatoms on this region preferentially migrate toward the edge of metal stripes with maximum stress. This result may show a possible way for the interconnection between randomly distributed self-assembled quantum dots and metal stripes.

Si delta-doping in the GaN layer has been successfully demonstrated by low-pressure metalorganic chemical vapor deposition at a growth temperature of 1040°C. Si delta-doping concentration increases and then decreases with an increase in delta-doping time. This indicates that delta-doping concentration is limited by the desorption process owing to much higher thermal decomposition efficiency of silane at high growth temperatures of GaN. In addition, it was observed that the use of a post-purge step in the ammonia ambient reduces Si delta-doping concentration. From capacitance-voltage measurement, a sharp carrier concentration profile with a full-width at half maximum of 4.1 nm has been achieved with a high peak concentration of 9.8 x 1018 cm-3