Applied Multivariate Statistical Analysis

Härdle, Wolfgang Karl; Simar, Léopold

doi:10.1007/978-3-642-17229-8

Cited by 345 publications

(312 citation statements)

References 0 publications

Supporting

Mentioning

289

Contrasting

Unclassified

Order By: Relevance

“…The terms 'strong', 'moderate', and 'weak' as applied to factor loadings refer to absolute loading values of C0.75, 0.75-0.50 and 0.50-0.30, respectively (Liu et al 2003). The factor scores were computed by the regression method which has the advantage of being the simplest technique and is easy to apply (Hardle and Simar 2007). Dalton and Upchurch (1978) have shown that factor scores for each sample could be related to intensity of the process described by the factor.…”

Section: Factor Analysismentioning

confidence: 99%

“…The priori groups correspond to natural classifications or groups known from history or prior knowledge and also could have been formed by a cluster analysis performed on past data (Hardle and Simar 2007). In this study, the Fisher's discriminant method was used.…”

Section: Discriminant Analysismentioning

confidence: 99%

“…In this study, the Fisher's discriminant method was used. Fisher's suggestion was to find the linear combination of datas which maximizes the ratio of the betweengroup-sum of squares to the within-group-sum of squares (Hardle and Simar 2007). The linear discriminant functions (LDF) is given by Eq.…”

Section: Discriminant Analysismentioning

confidence: 99%

See 2 more Smart Citations

Identifying the hydrochemical characteristics of rivers and groundwater by multivariate statistical analysis in the Sanjiang Plain, China

et al. 2014

View full text Add to dashboard Cite

Two multivariate statistical technologies, factor analysis (FA) and discriminant analysis (DA), are applied to study the river and groundwater hydrochemistry and its controlling processes in the Sanjiang Plain of the northeast China. Factor analysis identifies five factors which account for 79.65 % of the total variance in the dataset. Four factors bearing specific meanings as the river and groundwater hydrochemistry controlling processes are divided into two groups, the ''natural hydrochemistry evolution'' group and the ''pollution'' group. The ''natural hydrochemistry evolution'' group includes the salinity factor (factor 1) caused by rock weathering and the residence time factor (factor 2) reflecting the groundwater traveling time. The ''pollution'' group represents the groundwater quality deterioration due to geogenic pollution caused by elevated Fe and Mn (factor 3) and elevated nitrate (NO 3 -) introduced by human activities such as agriculture exploitations (factor 5). The hydrochemical difference and hydraulic connection among rivers (surface water, SW), shallow groundwater (SG) and deep groundwater (DG) group are evaluated by the factor scores obtained from FA and DA (Fisher's method). It is showed that the river water is characterized as low salinity and slight pollution, and the shallow groundwater has the highest salinity and severe pollution. The SW is well separated from SG and DG by Fisher's discriminant function, but the SG and DG can not be well separated showing their hydrochemical similarities, and emphasize hydraulic connections between SG and DG.

show abstract

Section: Factor Analysismentioning

confidence: 99%

Section: Discriminant Analysismentioning

confidence: 99%

See 1 more Smart Citation

Identifying the hydrochemical characteristics of rivers and groundwater by multivariate statistical analysis in the Sanjiang Plain, China

et al. 2014

View full text Add to dashboard Cite

show abstract

“…Some well-known similarity coefficients can be found in the literature [37] such as Jaccard coefficient (λ = 1) and Dice (λ = 0.5). On the other hand, the dissimilarity between two nodes can be also represented as a distance.…”

Section: Broadcasting In Disaster Response Scenarios As a Multi-objecmentioning

confidence: 99%

Multi-objective performance optimization of a probabilistic similarity/dissimilarity-based broadcasting scheme for mobile ad hoc networks in disaster response scenarios

et al. 2013

View full text Add to dashboard Cite

decentralized communications quickly and using common devices like mobile phones. Broadcasting is the main mechanism used to disseminate information in all-to-all fashion in ad hoc networks. The objective of this paper is to optimize a broadcasting scheme based on similarity/dissimilarity coefficient designed for disaster response scenarios through a multi-objective optimization problem in which several performance metrics such as reachability, number of retransmissions and delay are optimized simultaneously.

show abstract

“…3 представлены оценки плотности распределения Парзена-Розен-блатта [10] (при построении оценок использовалось гауссово ядро) для пред-ставителей трех возрастных групп дуба монгольского, находящихся в зонди-рующем окне с минимальным значением коэффициента корреляции (оценка плотности получена для проекций центров объектов на найденное оптималь-ное направление n). Разбиение на три возрастные группы проводилось по диаметру стволов деревьев: интервал возможных диаметров образцов, попав-ших в окно, был разбит на три равные части; в соответствии с разбиением формировались три группы старшая, средняя и младшая.…”

unclassified