ASSESSMENT OF INFLUENCE IN CANONICAL CORRELATION ANALYSIS

By expanding the perturbation of covariance matrix in the powers of er-ror term,the influence functions for five canonical measurements in CCA are devel-oped and three sample versions are given.For generalized correlation coefficient p_z,the influence function is a quadratic form of r.v.z,and its distribution is considered.A practical example iUustrates the utility of the proposed influence functions.

ASSESSMENT OF LOCAL INFLUENCE FOR GENERALIZED CORRELATION COEFFICIENTS

In this paper, we study local influence analysis for Zhang's generalized correlation coefficients and Hotelling's generalized correlation coefficient by using approach. of local influence analysis suggested by Shi (1991), i.e., generalized influence function (GIF) and generalized Cook distance (GCD). An example is given to illustrate our results.

The Influence Function of the Correlation Indexes in a Two-by-Two Table

In this paper we examine 5 indexes (the two Yule’s indexes, the chi square, the odds ratio and an elementary index) of a two-by-two table, which estimate the correlation coefficient ρ in a bivariate Bernoulli distribution. We will find the compact expression of the influence functions, which allow the quantification of the effect of an infinitesimal contamination of the probability of any pair of attributes of the bivariate random variable distributed according to the above-mentioned model. We prove that the only unbiased index is the chi square. In order to determine the indexes, which are less sensitive to contamination, we obtain the expressions of three synthetic measures of the influence function, which are the maximum contamination (gross sensitivity error), the mean square deviation and the variance. These results, even if don’t allow a definitive assessment of the overall optimum properties of the five indexes, as not all of them are unbiased, nevertheless they allow to appreciating the synthetic entity of the effect of the contaminations in the estimation of the parameter ρ of the bivariate Bernoulli distribution.

Correlation Analysis between Fruit Cracking and Influencing Factors in Apricot

In this study, 14 representative apricot cultivars in the production were selected as the experimental materials, and their fruit cracking characteristics, as well as the correlations between fruit cracking and influencing factors （e.g., pedcarp structure, mineral elements contents, DW/FW ratio and soluble sugar content） were analyzed to provide some reference for systematic study on fruit cracking mecha- nism of apricot. The results showed the cultivars with small orderly-and compactlyarranged epidermal cells were difficult to crack, while the cultivars with big disorderly-and loosely-arranged epidermal cells were easy to crack. There was no significant correlation between pericarp thickness and cracking index. The correlations between cracking and mineral elements contents of apricot fruit were in the order as Ca 〉 Zn 〉 Mn 〉 Fe 〉 K 〉 Mg 〉 Cu. The cracking index of apricot fruit was significantly negatively correlated with Ca content, was weakly correlated with Zn and Mn contents, and was uncorrelated with Fe, K, Mg and Cu contents. Ca deficiency was the main factor affecting the fruit cracking in apricot. Under the same conditions, the higher the water content is, the lower the cracking index is; and the higher the soluble sugar content is, the higher the cracking index is.

Analysis of Factors Influencing Comprehensive Productivity of Agriculture in Henan Province on the Basis of Grey Correlation

According to the data from Henan Statistical Yearbook from 2002 to 2008, from production capital, production conditions, labour inputs and financial support, this paper selects 11 variables influencing comprehensive productivity of agriculture in Henan Province. Through calculation and analysis of grey correlation of variables and comprehensive productivity of agriculture, this paper determines the impact of different variables on comprehensive productivity of agriculture. The results show that the agricultural capital has become the most important factor influencing comprehensive productivity of agriculture in Henan Province, while the impact of production conditions, labour inputs and financial support on comprehensive productivity of agriculture in Henan Province diminishes in turn. Corresponding countermeasures and suggestions are put forward to promote the sustainable development of comprehensive productivity of agriculture in Henan Province as follows: strengthen agricultural financial system building, and ensure agricultural production expenditure; scientifically arrange allocation of agricultural resources, and improve agricultural production conditions; carry out training of agricultural skills, and elevate the quality of agricultural labour forces; increase financial expenditure for agricultural production, and optimize financial expenditure structure.

Analysis of the Influence Factors of Grain Supply-Demand Gap in China

Based on the analysis of the grain supply and demand gap’s current situation in China, this paper establishes an indicator system for the influence factors of grain supply and demand gap. Then this paper calculates the correlation degree between the main grain varieties’ supply and demand gap and its influence factors. The results show that sown area and unit yield have the greatest impact on wheat supply and demand gap;per capita disposable income and unit yield have the greatest impact on corn supply and demand gap;per capita disposable income and agricultural mechanization level have the greatest impact on the supply and demand gap of soybean and rice. From the analysis results, we can obtain the difference between the factors affecting the grain supply and demand gap, and provide a certain theoretical basis and new ideas for the balance of grain supply and demand in China.

Influence of Multi-component Failure on Mechanical Product Design Change

Design change is one of the most common activities in design and development of mechanical products, and the change will affected by the design and maintenance of the products. It is not uncommon that multiple components might run out of order simultaneously during the use of a product, which would have significant influences on the product performance. Such kind of multi-component failure must be considered early in the product design stage, therefore it is necessary to study multiple design changes incurred by multi-component failure. Existing studies on the design change only focus on single change and ignore multiple simultaneous changes. To address this problem, a change influence network model is establishes, and the propagation path of multiple changes on the change influence network is described according to the change propagation characteristics of mechanical products. By analyzing the degree of dependence between the nodes and the absorption situations of nodes to change propagation, the possible paths of change propagation on the change influence network are determined, and based on which, calculation of the change influence is carried out, which provides decision-making support for product designers. Finally, the feasibility of the proposed methodology is demonstrated by taking a household juicer for case study.

Electricity Price Influence Factors Analysis Using Stochastic Matrix for Real-Time Electricity Price Forecasting

Real-time electricity price( RTEP) influence factor extraction is essential to forecasting accurate power system electricity prices. At present,new electricity price forecasting models have been studied to improve predictive accuracy,ignoring the extraction and analysis of RTEP influence factors. In this study,a correlation analysis method is proposed based on stochastic matrix theory.Firstly, an augmented matrix is formulated, including RTEP influence factor data and RTEP state data. Secondly, data correlation analysis results are obtained given the statistical characteristics of source data based on stochastic matrix theory.Mean spectral radius( MSR) is used as the measure of correlativity.Finally,the proposed method is evaluated in New England electricity markets and compared with the BP neural network forecasting method. Experimental results show that the extracted index system comprehensively generalizes RTEP influence factors,which play a significant role in improving RTEP forecasting accuracy.

Study on Water Injection Formula by Grey Correlation Method for Offshore Water Flooding Reservoir

Based on the development of the offshore water drive reservoir, the determination of reasonable water injection of monolayer and single well affects the distribution of remaining oil and development effect of oil field water flooding, so this paper, using the grey correlation analysis method, studies the main factors influencing the dividing coefficient of water-injection and weight. This method takes into account the effective thickness, permeability, injection-production well spacing, the number of injection-production wells, and the direction of formation coefficient factors such as the influence of the dividing coefficient for injection. The formula for water-injection multi-factor controlled is established, at the same time, applying water injection formula of this article and the conventional on injection allocation of A-platform offshore injection wells. And the difference of single-well injection-allocation quantity in layers is analyzed. Based on a platform offshore, layers exist in the process of water flooding and the problems such as the contradiction between the layers, put forward the necessity of layered water injection through this article to determine water injection formula and injection allocation on a platform of injection wells by implementing effect which can be found. A platform for water flooding development of water cut under control, natural decline rate reduced to verify the reliability and practicability of the formula to determine injection in this paper. It has certain guiding significance to the offshore for the water-injection development oilfield.

Improving the understanding of the influencing factors on sea level based on wavelet coherence and partial wavelet coherence

Relationship between sea level change and a single climate indicator has been widely discussed.However,few studies focused on the relationship between monthly mean sea level(MMSL)and several key impact factors,including CO_(2) concentration,sea ice area,and sunspots,on various time scales.In addition,research on the independent relationship between climate factors and sea level on various time scales is lacking,especially when the dependence of climate factors on Nino 3.4 is excluded.Based on this,we use wavelet coherence(WC)and partial wavelet coherence(PWC)to establish a relationship between MMSL and its influencing factors.The WC results show that the influence of climate indices on MMSL has strong regional characteristics.The significant correlation between Southern Hemisphere sea ice area and MMSL is opposite to that between Northern Hemisphere sea ice area and MMSL.The PWC results show that after removing the influence of Nino 3.4,the significant coherent regions of the Pacific Decadal Oscillation(PDO),Dipole Mode Index(DMI),Atlantic Multidecadal Oscillation(AMO),and Southern Oscillation Index(SOI)decrease to varying degrees on different time scales in different regions,demonstrating the influence of Nino 3.4.Our work emphasizes the interrelationship and independent relationship between MMSL and its influencing factors on various time scales and the use of PWC and WC to describe this relationship.The study has an important reference significance for selecting the best predictors of sea level change or climate systems.

缺血性脑卒中患者伤残接受度的相关影响因素分析

目的 研究影响缺血性脑卒中患者伤残接受度的相关影响因素。方法 选取2021年1月至2022年10月河北医科大学第二医院神经内科住院的缺血性脑卒中患者696例,采用一般资料调查表、伤残接受度量表、自我管理能力量表、社会支持量表及焦虑自评量表进行调查。结果 缺血性脑卒中患者伤残接受度量表得分为(65.63±11.30)分。多元线性回归结果显示,影响缺血性脑卒中患者伤残接受度的主要因素为文化程度、经济水平、对病情了解情况(P<0.05),自我管理能力与其伤残接受度呈正相关(r=0.557,P<0.01),社会支持度与其伤残接受度呈正相关(r=0.363,P<0.05),焦虑与其伤残接受度呈负相关(r=-0.665,P<0.01)。结论 缺血性脑卒中患者伤残接受度接受水平较低,与患者自我管理能力、社会支持呈正相关,与焦虑呈负相关,医务人员应给予重视,及时对患者进行心理干预,促进伤残接受度水平的提高,改善其身心健康。

陆浑灌区实际蒸散发影响因素分析

实际蒸散发是水文循环的关键环节,分析灌区实际蒸散发及其影响因素对灌区水资源的高效利用和农业高质量发展具有重要意义。然而,目前蒸散发的影响因素研究在确定主要因素时往往采用解释力较差的传统统计学方法,在相关性分析时忽略了蒸散发与其影响因素在空间上的相关性。因此利用改进的随机森林模型确定实际蒸散发的主要影响因素,并通过岭回归模型和地理加权回归模型探究实际蒸散发与其影响因素的时空相关关系。结果表明:(1)在灌溉期,地表净辐射、平均气温、叶面积指数和实际水汽压是实际蒸散发的主要影响因素;在非灌溉期,地表净辐射、平均气温、风速和日照时间是实际蒸散发的主要影响因素。实际蒸散发在一定程度上代表了灌区的作物耗水量。因此,灌区作物耗水在灌溉期和非灌溉期的影响作用有一定的差异。(2)在时间上,风速与实际蒸散发为负相关关系且呈显著负相关(P<0.05),其余影响因素与实际蒸散发均为正相关关系且呈显著正相关(P<0.05);在空间上,除风速与实际蒸散发在大部分区域呈负相关关系,其余影响因素都与实际蒸散发在大部分区域呈正相关关系。因此,除风速外,其余影响因素对灌区作物耗水在大部分区域都为正向促进作用。

人工授精女性痛苦表露状况及影响因素分析

目的:了解人工授精女性痛苦表露状况并分析其影响因素。方法:选择2021年4月-2023年4月本院收治的人工授精不孕症女性患者148例,使用一般资料调查问卷、痛苦表露指数量表(DDI)、领悟社会支持量表(PSSS)、Locke-Wallace婚姻调试量表(MAT)进行调查,以Pearson相关系数模型分析PSSS评分、MAT评分与人工授精女性痛苦表露水平的关系,并以元线性回归模型分析人工授精女性痛苦表露影响因素。结果:本次共发放问卷148份,有效回收138份,有效率为93.2%。138例女性DDI评分35.25±5.48分,低水平占比14.5%、中等水平占比73.2%、高水平占比12.3%;PSSS评分53.15±9.67分,MAT评分99.56±12.21分。单因素分析显示,不同年龄、家庭人均月收入、居住地、是否为夫精人工授精、人工授精次数的人工授精女性痛苦表露评分存在差异,Pearson相关分析PSSS、MAT评分与人工授精女性DDI评分呈正相关(均P<0.001);多元线性回归分析,家庭人均月收入、是否为夫精人工授精、PSSS评分、MAT评分对人工授精女性痛苦表露水平有正向影响,人工授精次数对人工授精女性痛苦表露水平有负向影响(均P<0.05)。结论:本次调查的人工授精女性痛苦表露处于中等水平,家庭人均月收入、是否为夫精人工授精、人工授精次数、社会支持、婚姻亲密关系等是其重要影响因素,临床应高度重视并给予针对性护理干预,以提升人工授精女性痛苦表露水平,促进身心健康。

妊娠期妇女抑郁情绪影响因素及其与人格特征的相关性

目的探讨妊娠期妇女抑郁情绪影响因素,分析抑郁情绪与人格特征的相关性。方法应用一般资料调查表、爱丁堡产后抑郁量表(EPDS)、艾森克人格问卷简式量表中国版(EPQ-RSC)、妊娠压力量表(PPS)对246名妊娠期妇女进行调查,采取Pearson相关分析评估抑郁情绪与人格特征和妊娠压力的相关性,采取Logistic回归分析评估妊娠期妇女抑郁情绪的影响因素。结果完成调查的242名妊娠期妇女中,共有36名孕妇EPDS评分≥13分(抑郁情绪组),206名孕妇EPDS评分<13分(正常组),抑郁情绪检出率为14.88%。抑郁情绪组妊娠期妇女精神质、神经质人格评分,PPS总分及各维度压力源评分均高于正常组,外倾性人格评分低于正常组(P<0.01)。妊娠期妇女抑郁情绪与精神质、神经质人格评分,PPS总分及各维度压力源评分均呈显著正相关,与外倾性人格评分呈显著负相关(P<0.01)。外倾性、精神质、神经质人格评分,PPS总分及文化程度、家庭人均月收入、妊娠意愿、睡眠情况、夫妻关系均是妊娠期妇女抑郁情绪的影响因素(P<0.05或0.01)。结论妊娠期妇女易产生抑郁情绪,且与人格特征、妊娠压力密切相关,医护人员应重视妊娠期妇女抑郁情绪相关影响因素评估,制定科学干预措施,改善抑郁情绪。

新入职护士时间管理倾向与临床实践行为的调查研究

目的探讨新入职护士时间管理倾向与临床实践行为的现状、相关性及影响因素。方法于2022年1月选取九江市的市级、县级5家医院的145名新入职护士作为调查对象,采用问卷调查的方式,对护士进行一般人口学特征6项,时间管理倾向量表3个维度及临床实践行为3个维度的调查,比较不同人口学特征护士的时间管理倾向、临床实践行为,采用Pearson检验分析护士时间管理倾向与临床实践行为的相关性,采用多元线性回归分析临床实践行为的影响因素。结果145名护士的时间管理倾向量表总分为(3.74±0.62)分,其中时间价值感、时间监控观、时间效能感得分较高。临床实践行为评分总分为(4.15±0.64)分,其中认知、情感、操作技能得分较高。不同学历、生源地、所在医院级别、是否担任班干部的护士时间管理倾向评分比较,不同学历、生源地、所在医院级别的护士临床实践行为评分比较,差异有统计学意义(P<0.05)。时间管理倾向量表3个维度评分及总分,与临床实践行为3个维度评分及总分呈正相关(P<0.05)。多元线性回归分析显示,护士的学历、生源地、医院级别、时间管理倾向3个维度评分及总分为临床实践行为的影响因素(P<0.05)。结论新入职护士时间管理倾向与临床实践行为处于较高水平,而且呈现正相关,其中生源地、市级医院、时间管理倾向为临床实践行为的影响因素,因此需加强时间管理技巧的培训,提高护士的临床实践行为,研究价值较高。

基于抑郁程度、心理痛苦、自理能力分析的脑卒中后抑郁患者自杀倾向的影响因素探讨

目的调查脑卒中后抑郁(PSD)患者自杀倾向现况,探究基于抑郁程度、心理痛苦、自理能力进一步分析自杀倾向的影响因素。方法选取2021年1月—2023年1月神经内科住院治疗的PSD患者165例,采用MINI自杀筛选问卷评分、健康状况问卷抑郁量表(PHQ-9)评分、心理痛苦温度计(DT)评分、Barthel指数(BI)评分对PSD患者自杀倾向、心理痛苦及病情程度进行评估,并通过一般资料调查问卷收集患者一般情况,采用Pearson法分析MINI评分与PHQ-9评分、DT评分、BI评分的相关性,采用多元线性回归分析PSD患者自杀倾向的影响因素,采用受试者工作特征(ROC)曲线评价各因素及预测模型预测PSD患者自杀倾向的敏感度、特异度,进一步运用部分标准化方法计算各变量重要性。结果165例PSD患者MINI评分(6.65±2.38)分,其中无风险、低风险、中风险、高风险分别占12.12%、24.24%、29.70%、33.94%;PSD患者PHQ-9评分(10.58±3.12)分,DT评分(5.33±2.65)分,BI评分(56.38±6.75)分。MINI评分与PHQ-9评分、DT评分呈正相关,与BI评分呈负相关(P<0.01);多元线性回归分析显示:经济收入、PHQ-9评分、DT评分、简明健康状况量表(SF-36)评分、BI评分、对预后有信心、对治疗效果满意、自杀未遂经历是PSD患者自杀倾向的影响因素(P<0.01);重要性排序前五的因素依次为对预后有信心、BI评分、DT评分、PHQ-9评分、自杀未遂经历;ROC曲线显示各因素及预测模型均具有良好的预测价值。结论对预后有信心、BI评分、DT评分、PHQ-9评分、自杀未遂经历是PSD患者自杀倾向重要性排名前五的影响因素,其余因素有经济收入、SF-36评分、对治疗效果满意,临床可以此为依据展开有针对性、有侧重点的防控措施,以降低PSD患者自杀倾向。

变量之间的影响关系和多重影响因素的共同作用类型

探讨研究变量之间的关系是心理学和其他社科领域实证研究的主要工作。基于问卷调查的实证文章多数涉及变量之间的“影响关系”,但这个概念未有明确界定,经常被人理解为相关关系或者因果关系,无论哪种情况都会带来问题,尤其在中介效应研究中更为突出。本文将影响关系定义为有方向性的相关关系,从外延和内涵上揭示了相关、影响和因果三者之间的关系;提供了多种途径为影响关系建模找到理据;讨论了一个结果变量的多重影响因素的共同作用类型。本文将为问卷研究中的变量之间关系研究提供理论支撑。

糖尿病专科护士核心能力影响因素的结构方程分析

目的通过相关性分析和结构方程模型阐明影响糖尿病专科护士核心能力的影响因素,以期为糖尿病专科护士的管理和培养提供理论依据。方法选取广东省护理学会2016年—2021年培养的糖尿病专科护士为研究对象。采用自制的核心能力相关因素问卷(包括糖尿病护理及教育相关行为、糖尿病护理科研相关行为、继续教育相关行为、工作环境相关支持共4个维度)、糖尿病专科护士核心能力问卷对研究对象进行调查。共计发放问卷213份,回收有效问卷213份,有效回收率100%。采用相关性分析和结构方程模型对糖尿病专科护士核心能力的影响因素进行分析。结果广东省糖尿病专科护士核心能力总体处于中等偏上水平,得分较低者为科研能力20.02±6.86分,得分最高者为健康教育能力70.25±9.45分。糖尿病护理及教育相关行为、糖尿病护理科研相关行为、继续教育相关行为、工作环境相关支持与糖尿病专科护士核心能力之间具有相关性(r=0.156~0.764,P<0.05);结构方程模型显示,护理及教育相关行为对核心能力的影响最大(0.25),其次为科研相关行为(0.23),继续教育相关行为(0.13),工作环境相关支持对核心能力的影响最小(0.07)。科研相关行为、继续教育相关行为、工作环境相关支持可通过护理及教育相关行为的中介作用间接影响糖尿病专科护士核心能力。结论护理管理者应该重视糖尿病专科护士护理及教育相关行为,促进其科研相关行为,鼓励继续教育相关行为并提供工作环境相关支持来提高糖尿病专科护士核心能力。

大学生自主在线学习意识及在线学习效果的实证研究

在新冠疫情背景下在线学习的常态化发展,大学生的自主在线学习意识和学习效果已然成为评价学习质量和教育质量的观察指标。该研究旨在通过对336名大学生进行自主在线学习意识、学习效果及两者关系的研究分析,为在线学习意识和效果的提高提出相关建议。采用曼-惠特尼U检验和Kruskal-Wallis检验分析进行在线学习意识、学习效果的现状及差异性分析;采用多因素线性回归分析在线学习意识和学习效果的影响因素;在线学习意识和学习效果的相关性检验采用Spearman相关性分析。研究发现影响大学生自主在线学习意识和学习效果的因素包括专业、年级、学习时间、学习频率及网络环境。建议学校针对不同专业、年级学生合理有计划地安排自主在线学习时间和频率,优化在线学习网络环境和在线学习资源,进而提高自主在线学习意愿和学习效果。

新质生产力视域下中国高校科技创新力评价及影响因素研究

习近平总书记立足于我国经济发展时代特征提出新质生产力概念,并作出一系列重要论述。新质生产力是在新技术革命推动下,以信息技术、人工智能为引擎,以科技创新为驱动力的生产力形态,是高质量发展的强劲推动力、支撑力。高校作为国家创新体系的重要组成部分,在基础研究、应用研究、成果转化等方面发挥着重要作用,取得了一批重大科技成果,但还存在着诸多不足,面临不少困难和问题。在习近平总书记关于新质生产力重要论述的指引下,准确把握高校科技创新的重要地位和独特优势,客观评价我国不同地区高校科技创新力发展水平,分析存在的问题,提出对策建议,具有极其重要的意义。文章采用TOPSIS法测算我国31省(自治区、直辖市)高校科技创新力指数,基于空间自相关理论探究我国高校科技创新力指数的空间关联特征,运用泰尔指数分析我国高校科技创新力指数的空间差异性,运用地理探测器原理研究我国高校科技创新力的影响因素及其交互作用。研究结果发现:我国31省(自治区、直辖市)高校科技创新力整体还不强,创新力指数的空间分布不均衡,东部地区明显高于中部和西部地区;空间自相关性显著,上海、江苏、广东、浙江等指数较高省份呈现“高-高”集聚特征,甘肃、海南、新疆、西藏等指数较低省份呈现“低-低”集聚特征;区域间差异明显,东部和西部地区域内差异较中部地区显著,东部地区内部差异逐渐缩小,中部地区内部差异先增大后缩小,西部地区内部差异有所增大;影响我国高校科技创新力发展的主要因素是人力资源、经费投入等,且区域间投入存在明显差异,中部和西部地区投入不足,同时还受区域对外开放和技术需求等因素影响;内部因素交互作用强度大于外部因素,交互作用强度较大的是教学与科研人员数、科技经费当年拨入、成果应用及科技服务项目数、R&D成果应用及科技服务项目经费当年拨入、信息化水平,且信息化水平发挥的作用越来越大。根据研究结论,以习近平总书记关于新质生产力重要论述为指引,从创新投入、资源配置、资源共享、政产学研合作等方面提出对策建议。