Age related variation of salivary testosterone values in healthy Japanese males

Objective. We examined age associated variation in salivary testosterone values among Japanese males as well as anthropometric measurements.

Methods. Salivary samples were collected in pretreated sodium azide treated tubes. The first series: 15–79-year-old males (n = 99); two morning and two evening samples were collected at home for two days. The second series: 90-year-old males (n = 29); one morning sample was collected. Testosterone values were determined using an iodine125-based radioimmunoassay kit modified for saliva.

Results. Results show 1) a significant decrease in salivary testosterone values from 20s to 40s and older, 2) no significant decline after 40 through 90 years old, 3) no significant age-related differences in the degree of intraindividual diurnal fluctuation across age groups of 40–70s, and 4) higher BMI is associated with the lower salivary testosterone among 40–70s.

Conclusions. These results suggest that neither a constant decrease of salivary testosterone values or markedly reduced intraindividual fluctations are universal aspects of aging. Older males may maintain relatively high testosterone levels compared to younger men and a relatively ‘robust’ neuroendocrinological system.     

Lifetime Assessment by Intermittent Inspection under the Mixture Weibull Power Law Model with Application to XLPE Cables

This paper proposes a new treatment for electrical insulation degradation. Some types of insulation which have been used under various circumstances are considered to degrade at various rates in accordance with their stress circumstances. The cross-linked polyethylene (XLPE) insulated cables inspected by major Japanese electric companies clearly indicate such phenomena. By assuming that the inspected specimen is sampled from one of the clustered groups, a mixed degradation model can be constructed. Since the degradation of the insulation under common circumstances is considered to follow a Weibull distribution, a mixture model and a Weibull power law can be combined. This is called The mixture Weibull power law model. By using the maximum likelihood estimation for the newly proposed model to Japanese 22 and 33kV insulation class cables, they are clustered into a certain number of groups by using the AIC and the generalized likelihood ratio test method. The reliability of the cables at specified years are assessed. This revised version was published online in July 2006 with corrections to the Cover Date.     

Joint Modeling of Survival and Longitudinal Ordered Data Using a Semiparametric Approach

Medical research frequently focuses on the relationship between quality of life (QoL) and survival time of subjects. QoL may be one of the most important factors that could be used to predict survival, making it worth identifying factors that jointly affect survival and QoL. We propose a semiparametric joint model that consists of item response and survival components, where these two components are linked through latent variables. Several popular ordinal models are considered and compared in the item response component, while the Cox proportional hazards model is used in the survival component. We estimate the baseline hazard function and model parameters simultaneously, through a profile likelihood approach. We illustrate the method using an example from a clinical study.     

Readouts for echo-state networks built using locally regularized orthogonal forward regression

Echo state network (ESN) is viewed as a temporal expansion which naturally give rise to regressors of various relevance to a teacher output. We illustrate that often only a certain amount of the generated echo-regressors effectively explain the teacher output and we propose to determine the importance of the echo-regressors by a joint calculation of the individual variance contributions and Bayesian relevance using the locally regularized orthogonal forward regression (LROFR). This information can be advantageously used in a variety of ways for an analysis of an ESN structure. We present a locally regularized linear readout built using LROFR. The readout may have a smaller dimensionality than the ESN model itself, and improves robustness and accuracy of an ESN. Its main advantage is ability to determine what type of an additional readout is suitable for a task at hand. Comparison with PCA is provided too. We also propose a radial basis function (RBF) readout built using LROFR, since flexibility of the linear readout has limitations and might be insufficient for complex tasks. Its excellent generalization abilities make it a viable alternative to feed-forward neural networks or relevance-vector-machines. For cases where more temporal capacity is required we propose well studied delay&sum readout.     

Variable selection via the weighted group lasso for factor analysis models

We consider the problem of selecting variables in factor analysis models. The $L_1$ regularization procedure is introduced to perform an automatic variable selection. In the factor analysis model, each variable is controlled by multiple factors when there are more than one underlying factor. We treat parameters corresponding to the multiple factors as grouped parameters, and then apply the group lasso. Furthermore, the weight of the group lasso penalty is modified to obtain appropriate estimates and improve the performance of variable selection. Crucial issues in this modeling procedure include the selection of the number of factors and a regularization parameter. Choosing these parameters can be viewed as a model selection and evaluation problem. We derive a model selection criterion for evaluating the factor analysis model via the weighted group lasso. Monte Carlo simulations are conducted to investigate the effectiveness of the proposed procedure. A real data example is also given to illustrate our procedure. The Canadian Journal of Statistics 40: 345–361; 2012 © 2012 Statistical Society of Canada     

Socio-material arrangements of impoverished youth in Japan: historical and critical perspectives on neoliberalization

In the past decade, in Japan, a new light has been cast on youth poverty in the context of the middle-class impoverishment caused by the advance of neoliberalization. Thus, the socio-material arrangements, that is, the configuration of people and artifacts pertaining to poor Japanese children, was explored. The authors conducted a field survey in Tokyo’s Adachi Ward, where poverty and the problems it brings have become more serious, and then analyzed that data. In the analysis, we introduced a historical viewpoint to understand the changes in the socio-material arrangements of poor youth that have been brought about by neoliberalization.     

Rates of convergence for the k-nearest neighbor estimators with smoother regression functions

Let (X, Y  ) be a R^d×R-valued${\mathbb{R}}^d\times \mathbb{R}\mathrm{-}\mathrm{valued}$ random vector. In regression analysis one wants to estimate the regression function m(x)?E(Y|X=x)$m(x)?\mathbf{E}(Y|X=x)$ from a data set. In this paper we consider the rate of convergence for the k-nearest neighbor estimators in case that X   is uniformly distributed on [0,1]^d$[\mathrm{0,1}{]}^d$, Var(Y|X=x)$\mathbf{Var}(Y|X=x)$ is bounded, and m is (p, C)-smooth. It is an open problem whether the optimal rate can be achieved by a k  -nearest neighbor estimator for 1<p≤1.5$1<p\le 1.5$. We solve the problem affirmatively. This is the main result of this paper. Throughout this paper, we assume that the data is independent and identically distributed and as an error criterion we use the expected L₂ error.     

Creating facial animation of characters via MoCap data

We consider the problem of generating 3D facial animation of characters. An efficient procedure is realized by using the motion capture data (MoCap data), which is obtained by tracking the facial markers from an actor/actress. In some cases of artistic animation, the MoCap actor/actress and the 3D character facial animation show different expressions. For example, from the original facial MoCap data of speaking, a user would like to create the character facial animation of speaking with a smirk. In this paper, we propose a new easy-to-use system for making character facial animation via MoCap data. Our system is based on the interpolation: once the character facial expressions of the starting and the ending frames are given, the intermediate frames are automatically generated by information from the MoCap data. The interpolation procedure consists of three stages. First, the time axis of animation is divided into several intervals by the fused lasso signal approximator. In the second stage, we use the kernel k-means clustering to obtain control points. Finally, the interpolation is realized by using the control points. The user can easily create a wide variety of 3D character facial expressions by changing the control points.