成分向量分布的几个问题

成分统计分析的名著[1]中提出了一些看法,本文就其中三点提出了不同的意见,并给出了实例。(1)关于狄氏分布与独立性;(2)关于乘法逻辑分布的引入有无必要;(3)关于成分与总量的独立性。

用微探针型EDXRF技术研究南宋官窑瓷片的化学组成及其含义

本文报道了用ENGLE-II微探针能量色散型X荧光光谱(EDXRF)技术测定一批青瓷残片中的16种元素氧化物的含量,其残片来自浙江杭州地区南宋的郊坛窑和老虎洞窑。由主量元素判别方程Q值表明,郊坛和老虎洞青釉瓷均属官窑制品,与龙泉民窑产品有很大差别。从主成分统计分析看出,郊坛下官窑与老虎洞官窑的瓷釉成分有明显不同,但其胎质成分的分布却有部分的重叠。这显示两类同属官窑的制品,其釉料可有不同的矿物组合,但胎料有时会交流使用。从釉层、中间层、和胎层的Al2O3和CaO的含量变化数据表明,郊坛下官窑和老虎洞官窑的青瓷制品有厚度分别为120-140μm和230-240μm的中间层,反映出烧造工艺上有一定的差别。化学成分鉴定证实了考古工作者认为老虎洞窑实为史书记载的修内司官窑的观点。

众包环境下的任务定价因素研究

随着网络的高速发展,各式各样的劳务众包平台变得日益流行,"拍照赚钱"就是其中的一种,定价不合理会导致最后商品的检查失败,本文建立了动态定价模型,确立了最优折扣率来使得任务发包人和会员双方利益更均衡,以提高任务完成度,为企业提供各种有效的商业检查和信息搜索。

沉积物物源组成的定量估计方法

混合源沉积物中,各种物源物质的成分及其百分含量是地质研究的重要基础信息。本文提出用成分数据统计分析算法,分析处理沉积物样品的地球化学数据,定量提取沉积物中各物源端元的平均化学成分及其百分含量;将该方法用于已知端元成分和端元含量的实验样品数据,计算结果与实际非常吻合;将该方法用于某海区实测沉积物地球化学数据,成功地分离出了陆源、火山源、生物源物质成分和含量。该方法也适合于岩石学、环境污染研究等领域。

Statistical and causes analysis of storm surges along Tianjin coast during the past 20 years

Based on tidal data statistical analysis for 20 years of Tanggu Marine Environmental Monitoring Station from 1991 to 2010, we concluded that an average of nearly 10 days of 100 cm above water increase took place at Tianjin coast every year. The maximum high tide and average tide of Tianjin coast occurred in summer and autumn, and the maximum water increase also occurred in summer and autumn. Days with water increase more than 100 cm mostly occurred in spring, autumn and winter. Then we summarized the causes of coastal storm surge disaster in Tianjin based on astronomical tide factors, meteorological factors, sea level rise, land subsidence, and geographic factors, et al. Finally, we proposed storm surge disaster prevention measures.

Analysis of fatty acid composition of sea cucumber Apostichopus japonicus using multivariate statistics

Fatty acids(FAs) provide energy and also can be used to trace trophic relationships among organisms.Sea cucumber Apostichopus japonicus goes into a state of aestivation during warm summer months.We examined fatty acid profiles in aestivated and non-aestivated A.japonicus using multivariate analyses(PERMANOVA,MDS,ANOSIM,and SIMPER).The results indicate that the fatty acid profiles of aestivated and non-aestivated sea cucumbers differed significantly.The FAs that were produced by bacteria and brown kelp contributed the most to the differences in the fatty acid composition of aestivated and nonaestivated sea cucumbers.Aestivated sea cucumbers may synthesize FAs from heterotrophic bacteria during early aestivation,and long chain FAs such as eicosapentaenoic(EPA) and docosahexaenoic acid(DHA) that produced from intestinal degradation,are digested during deep aestivation.Specific changes in the fatty acid composition of A.japonicus during aestivation needs more detailed study in the future.

Statistical Monitoring of Chemical Processes Based on Sensitive Kernel Principal Components

The kernel principal component analysis （KPCA） method employs the first several kernel principal components （KPCs）, which indicate the most variance information of normal observations for process monitoring, but may not reflect the fault information. In this study, sensitive kernel principal component analysis （SKPCA） is proposed to improve process monitoring performance, i.e., to deal with the discordance of T2 statistic and squared prediction error SVE statistic and reduce missed detection rates. T2 statistic can be used to measure the variation di rectly along each KPC and analyze the detection performance as well as capture the most useful information in a process. With the calculation of the change rate of T2 statistic along each KPC, SKPCA selects the sensitive kernel principal components for process monitoring. A simulated simple system and Tennessee Eastman process are employed to demonstrate the efficiency of SKPCA on online monitoring. The results indicate that the monitoring performance is improved significantly.

Assessment of temporal and spatial variations in surface water quality using multivariate statistical techniques: A case study of Nenjiang River basin, China

Assessment of temporal and spatial variations in surface water quality is important to evaluate the health of a watershed and make necessary management decisions to control current and future pollution of receiving water bodies. In this work, surface water quality data for 12 physical and chemical parameters collected from 10 sampling sites in the Nenjiang River basin during the years(2012-2013) were analyzed. The results show that river water quality has significant temporal and spatial variations. Hierarchical cluster analysis(HCA) grouped 12 months into three periods(LF, MF and HF) and classified 10 monitoring sites into three regions(LP, MP and HP) based on the similarity of water quality characteristics. The principle component analysis(PCA)/factor analysis(FA) was used to recognize the factors or origins responsible for temporal and spatial water quality variations. Temporal and spatial PCA/FA revealed that the Nenjiang River water chemistry was strongly affected by rock/water interaction, hydrologic processes and anthropogenic activities. This work demonstrates that the application of HCA and PCA/FA has achieved meaningful classification based on temporal and spatial criteria.

Multivariate Statistical Process Monitoring of an Industrial Polypropylene Catalyzer Reactor with Component Analysis and Kernel Density Estimation

Abstract Data-driven tools, such as principal component analysis （PCA） and independent component analysis （ICA） have been applied to different benchmarks as process monitoring methods. The difference between the two methods is that the components of PCA are still dependent while ICA has no orthogonality constraint and its latentvariables are independent. Process monitoring with PCA often supposes that process data or principal components is Gaussian distribution. However, this kind of constraint cannot be satisfied by several practical processes. To ex-tend the use of PCA, a nonparametric method is added to PCA to overcome the difficulty, and kernel density estimation （KDE） is rather a good choice. Though ICA is based on non-Gaussian distribution intormation, .KDE can help in the close monitoring of the data. Methods, such as PCA, ICA, PCA.with .KDE（KPCA）, and ICA with KDE,（KICA）, are demonstrated and. compared by applying them to a practical industnal Spheripol craft polypropylene catalyzer reactor instead of a laboratory emulator.

Nonlinear Statistical Process Monitoring Based on Control Charts with Memory Effect and Kernel Independent Component Analysis

A novel nonlinear combination process monitoring method was proposed based on techniques with memo- ry effect （multivariate exponentially weighted moving average （MEWMA）） and kernel independent component analysis （KICA）. The method was developed for dealing with nonlinear issues and detecting small or moderate drifts in one or more process variables with autocorrelation. MEWMA charts use additional information from the past history of the process for keeping the memory effect of the process behavior trend. KICA is a recently devel- oped statistical technique for revealing hidden, nonlinear statistically independent factors that underlie sets of mea- surements and it is a two-phase algorithm., whitened kernel principal component analysis （KPCA） plus indepen- dent component analysis （ICA）. The application to the fluid catalytic cracking unit （FCCU） simulated process in- dicates that the proposed combined method based on MEWMA and KICA can effectively capture the nonlinear rela- tionship and detect small drifts in process variables. Its performance significantly outperforms monitoring method based on ICA, MEWMA-ICA and KICA, especially for lonu-term performance deterioration.

Authorship Identification of Classical Japanese Literature Using Quantitative Analysis

Singular authorship of＂The Tale of Genji,＂ which is the greatest accomplishment in classical Japanese literature, is doubtful In this study, we statistically analyze whether the author of Uji Jujo, which comprises the last ten chapters of the ＂Tale of Ganji,＂ is same as that of other chapters by using statistical analyses such as principal component analysis and random forests. The result of the analyses of word frequency shows that there is no obvious difference in word usage between the ten chapters and others. Thus, we conclude that it is highly possible that the chapters＇ author is same as that of others.

Integrated Application of Statistical Method Used in Predicting the Lanes＇ Traffic Volume of Non-detector Intersections

Based on the relationships between the lanes of signal-controlled intersections, we utilize the integration method of cluster analysis and stepwise regression and the integration method of cluster analysis and the principal component analysis method to predict the lanes＇ traffic volume of non-detector isolated controlled intersections. The results are examined by the real-time lanes＇ traffic volume data of the road network of Nanjing City. The problem of the lanes＇ traffic volume prediction of non-detector isolated signal-controlled intersections was resolved which can be widely used in urban traffic flow guidance and urban traffic control in cities.

Comparative study on main compounds in Ginkgo biloba L. base on mathematical statistics

In this paper,an HPLC-DAD-ELSD method was developed to determine main 20 components of Ginkgo biloba L.leaves from different ages and sources,including six flavonol glycosides,five terpene lactones and nine organic acids.Using statistics method and establishing relevant mathematics models,the measured data has proceeded correlation analysis,principal component analysis,and regression statistics and the results showed generality and specific characteristics.We defined p-hydroxybenzoic acid,catechinic,KRcG and ginkgolide A as characteristic indexes representing commonness and speciality of Ginkgo biloba L.leaf.The four characteristic indexes can reflect the quality of Ginkgo biloba L.leaf,and the intermal relations between them are significant.The contents of other compounds could define the quantity relation with characteristic markers.It simplified the approach of quality control,and provided a basis for quality control of Ginkgo biloba L.

Physical characteristics of Chinese Hakka

Using standard and internationally validated methods,86 anthropologic characteristics were determined in 650 male(305 from urban areas and 345 from rural areas) and 704 female(331 from urban areas,373 from rural areas) Chinese Hakka adults living in Guangdong and Jiangxi.The data were used to calculate 24 anthropologic indices,which were analyzed statistically.The physical characteristics of Hakka subjects were analyzed and compared with reference ethnic data.There were four main findings of this study.First,a small proportion of Hakka adults had an eye fold on the upper eyelid,but a large proportion had a mongoloid fold.The eye slits were narrow in most adults,had a medium nasal root height and straight bridges,and most of the external angles were prominent.The nasal base was upturned in most men.The distributions of the three types of nasal base in women were similar.The proportions of subjects with middle and high alae nasi heights were high and similar.Males with a maximum nostril diameter were mostly classified as transverse and oblique,while many women were classified as transverse and had relatively wide alae nasi.The round lobe type was the most common.Upper lip skin height was mostly classified as medium.Lips were classified as thin.The hair was black,eyes were brown,and the skin was yellowish.Second,the head length was long in male Hakka.The minimum frontal breadth,face breadth,lip height,and interocular breadth were similar to those of North-Asian populations.Meanwhile,head breadth,morphological facial height,nose breadth,mouth breadth,and nose height were similar to those of South-Asian populations.Head length was long in female Hakka.The minimum frontal breadth,face breadth,lip height,and interocular breadth were similar to those of North-Asian populations.Head breadth,nose breadth,and mouth breadth were similar to those of South-Asian populations.Third,the stature of male and female Hakka in urban and rural areas was classified as medium.The proportions of male and female Hakka classified as mesocephaly(length-breadth index of the head),hypsicephalic type,metriocephalic type(breadth-height index of the head) mesorrhiny,long trunk,subbrachyskelic type,broad shoulder breadth,and narrow distance between iliac crests were higher than those of other types.Finally,principal component analyses showed that the physical characteristics of Hakka were between those of South-Asian and North-Asian ethnic populations,but were generally closer to those of North-Asian populations in China.