An Adaptive Estimator of the Autocorrelation Coefficient in Regression Models with Autoregressive Errors

In this paper an adaptive estimator of the autocorrelation coefficient is constructed in regression models whose error variables follow a stationary autoregressive process of order 1. Examples of nonparametric, additive and semiparametric regression models are discussed.     

Application of chemical composition and infusion colour difference analysis to quality estimation of jasmine-scented tea

Infusion chemical composition, colour difference indicators and volatile constituents of seventeen jasmine‐scented tea samples and their correlation with sensory total quality score given by tea‐tasting panel were studied by techniques of HPLC, colour difference metre and gas chromatograph. Spearman's linear correlation analysis showed that concentration of nitrogen, caffeine, catechin, epicatechin, catechin gallate, total catechins, terpineol, nerol and infusion light‐black indicator ΔL, yellow‐blue indicator Δb, total colour difference indicator ΔE were significantly correlated with sensory total quality score, respectively. Regression of the total quality score upon twelve components extracted from thirty‐six tested indicators by principal component analysis could properly estimate total quality score of two jasmine‐scented tea samples which were not included in the seventeen samples used to construct the regression model.     

Hair mercury levels in relation to fish consumption in a community of the Moroccan Mediterranean coast

Coastal populations with high seafood consumption in the Mediterranean have a significant exposure to dietary methylmercury, and areas where environmental mercury pollution is an issue due to industrial activities are of special concern. The study was undertaken with the aim of assessing methylmercury exposure through fish consumption in a community of north Morocco and characterizing the relevant health risk. Concentrations of total mercury were determined in human hair, a biomarker of methylmercury exposure, and in locally consumed fish by cold vapour atomic absorption spectrophotometry. Based on consumption frequencies reported by the 108 subjects included in the study the weekly intake of methylmercury was estimated and biomarker data were evaluated in relation to the estimated intake and the sociodemographic characteristics of the population. Multiple regression analysis was employed for the interpretation of hair mercury data in relation to fish consumption frequency, gender and the age of individuals. Mercury concentrations in hair ranged from 0.22 to 9.56 µg g^-1 (geometric mean = 1.79 µg g^-1) and were closely related to fish intake. Fisherman and their families consumed fish three to five times per week and were the most exposed population subgroup. A high proportion of women of child-bearing age (50%) had relatively high levels of mercury in their hair (3.08-7.88 µg g^-1).     

二滩水电站水位计测值分析

运用数理统计的方法和最小二乘法的原理，对二滩水电站大坝３支水位计观测资料进行分析，求出回归方程并经实践检验，证明水位计是可以继续工作的。     

Physicochemical and sensory analysis of dry bean landraces in conjunction with multivariate analysis: an exploratory approach

Five landraces of Greek common dry bean (Phaseolus vulgaris L.), were analysed for thirteen physico‐chemical and eighteen sensory variables. Brightness, acceptance, odour and hardness were the most important sensory variables for the overall acceptability of the cooked product. Application of principal component analysis revealed a rather weak relationship between the sensory attributes and the physico‐chemical variables. However, a certain grouping of cultivars with corresponding properties [(PC1 vs. PC2, Kastoria with saltiness, hardness, grassiness, intensity, protein and metal; Byzitsa bushy with fat, odour and sourness; Ksanthi and Belestino with tenaciousness, chroma, pH and overall impression), (PC1 vs. PC3 Byzitsa bushy with odour and acridness; Ksanthi with metal, hardness and pH; Kastoria with colour, protein and intensity)] was made. On the contrary, cluster analysis showed a much more satisfactory grouping of dry bean varieties based on their geographical origin. The multiple regression equation explained 72.3% of the variation in total acceptance of dry beans. The relative importance of brightness, acceptance, smell and hardness in predicting the overall impression of dry beans was 13%, 13.7%, 26% and 10%, respectively. These results may indicate that smell, brightness and acceptance are the most important factors determining dry bean total acceptability, so effort should be directed at improving these attributes in an attempt to satisfy consumer experience.     

Demonstration Analysis of Relationship Between R&D Investment and GDP

To reveal the quantitative relationship between research and development (R&D) investment and gross domestic product (GDP) in China, we have demonstrated and analyzed the relationship between R&D investment and science and technology (S&T) progress, and based on a mount of S&T statistical data, have proceeded demonstration research of the relationship between R&D investment and GDP in China with Solow and vector auto regression (VAR) models. Cubic curve fitting and cross-correlation analysis of them with SPSS have shown that there is a strong synchronic relationship between R&D investment and GDP.     

Material Growth and Device Fabrication of GaN-Based Blue-Violet Laser Diodes

Studies on first GaN-based blue-violet laser diodes(LDs) in China mainland are reported.High quality GaN materials as well as GaN-based quantum wells laser structures are grown by metal-organic chemical vapor deposition method.The X-ray double-crystal diffraction rocking curve measurements show the fullwidth half maximum of 180″ and 185″ for (0002) symmetric reflection and (10-12) skew reflection,respectively.A room temperature mobility of 850cm2/(V·s) is obtained for a 3μm thick GaN film.Gain guided and ridge geometry waveguide laser diodes are fabricated with cleaved facet mirrors at room temperature under pulse current injection.The lasing wavelength is 405.9nm.A threshold current density of 5kA/cm2 and an output light power over 100mW are obtained for ridge geometry waveguide laser diodes.