Multivariate correlation analysis and its application in environmental analysis

Geiß, Sabine; Einax, Jürgen W.; Danzer, Klaus

doi:10.1016/0003-2670(91)87040-e

Cited by 12 publications

(11 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For many years, statistical methods have been applied extensively to analyze multivariate data sets in different fields of the earth sciences (Davis, 1986;Rock, 1988;Jackson, 1991;Christopherson and Hooper, 1992;Helsel and Hirsch, 1992;Brown, 1993). Multivariate geostatistical analysis deserves special attention because of the large number of topics encountered in the literature (Cooper and Istok, 1988a,b;Woodbury and Sudicky, 1991;Geiss et at., 1991;Istok et al, 1993;Wen and Kung, 1993). In these studies, multivariate data sets are considered as formed from regionalized variables, having continuities from point to point but varying in a complex manner, and they are evaluated by using semivariograms (a measure to describe spatial variability) and kriging (a weighted moving average technique) methods.…”

mentioning

confidence: 99%

“…In these studies, multivariate data sets are considered as formed from regionalized variables, having continuities from point to point but varying in a complex manner, and they are evaluated by using semivariograms (a measure to describe spatial variability) and kriging (a weighted moving average technique) methods. Tn recent years, several statistical applications have focused on topics related to environmental problems (Geiss et at., 1991;Melloul andCollin, 1991, 1992;Xhoffer et at., 1991;Soderstrom, 1992;Pohlmann, 1993;Zitko, 1994;Melloul, 1995).…”

mentioning

confidence: 99%

See 1 more Smart Citation

coastal aquifer by using correlation, principal component, and factor analyses

Karahanoğlu¹

1997

Water Environment Research

View full text Add to dashboard Cite

Hydrogeochemical characterization of the Erzin Plain coastal aquifer has been accomplished in this research to investigate the spatial and transient behavior of its water quality. The aquifer is located along the Mediterranean coast and forms one of the most productive aquifers in Turkey. Chemical analyses of groundwater samples collected from the aquifer during 1964–1968 (128 samples), May 1993 (30 samples), October 1993 (37 samples), and June 1994 (20 samples) constitute the available groundwater quality data set. The characteristics of the groundwater samples were evaluated statistically by correlation, principal component, and factor analyses to investigate interrelations among various groundwater quality and aquifer property parameters and to assess the occurrence of seawater intrusion. Application of principal component and factor analyses yielded two groups of variables indicative of seawater and freshwater. The basin is being contaminated by seawater, although it was not possible to locate the intrusion due to lack of sufficient data. The study emphasizes that management policies be reorganized for optimal use of the aquifer.

show abstract

mentioning

confidence: 99%

mentioning

confidence: 99%

coastal aquifer by using correlation, principal component, and factor analyses

Karahanoğlu¹

1997

Water Environment Research

View full text Add to dashboard Cite

show abstract

“…The measure proposed here is also an improvement on Geiss [4] by accounting for normalisation of the difference signal channels and use of negative eigenvalues. This measure can be adapted slightly for image region matching, to eliminate matches using regions with low contrast.…”

Section: Discussionmentioning

confidence: 99%

“…al. [4] proposed a multi-variate cross-correlation function: Define the image samples to be X{ and tji, both of dimension M. As before, assume that there is a set of matched pixels i = 1..7V from some appropriate image neighbourhood. Then, the mean pixel values in the neighbourhood are:…”

Section: Multi-variate Cross-correlationmentioning

confidence: 99%

Multi-Variate Cross-Correlation and Image Matching.

Fisher¹,

Oliver²

1995

Procedings of the British Machine Vision Conference 1995

View full text Add to dashboard Cite

This paper introduces the use of a multi-variate correlation function for region-based image matching and extends it to a modified crosscorrelation function that works well when matching image areas are required have the same intensity contrast. It also shows that the multivariate case is a straightforward generalisation of the monochrome image case. Experiments with both MRI and RGB colour imagery are shown, along with comparisons with the Euclidean, Manhatten and Loo matching metrics.

show abstract

“…The bivariate correlation analysis has been extensively discussed with numerous applications to different fields of study (Armstrong, 2019;Bareinboim et al, 2014;Brown et al, 2012;Geiss et al, 1991;Mukaka, 2012;Nguyen et al, 2013;Wang et al, 2017). Different techniques describing multivariate correlation analysis and their performance analysis have been discussed (Wang et al, 2017).…”

Section: Introductionmentioning

confidence: 99%

Robust Multivariate Correlation Techniques: A Confirmation Analysis using Covid-19 Data Set

Okwonu¹,

Ahad²,

Apanapudor³

et al. 2021

JST

View full text Add to dashboard Cite

Robust multivariate correlation techniques are proposed to determine the strength of the association between two or more variables of interest since the existing multivariate correlation techniques are susceptible to outliers when the data set contains random outliers. The performances of the proposed techniques were compared with the conventional multivariate correlation techniques. All techniques under study are applied on COVID-19 data sets for Malaysia and Nigeria to determine the level of association between study variables which are confirmed, discharged, and death cases. These techniques’ performances are evaluated based on the multivariate correlation (R), multivariate coefficient of determination (R^2), and Adjusted R^2. The proposed techniques showed R=0.99 and the conventional methods showed that R ranges from 0.44 to 0.73. The R^2 and the Adjusted R^2 for proposed methods are 0.98 and 0.97 while the conventional methods showed that R equals 0.53, 0.44, and 0.19 whereas Adjusted R^2 equals 0.52, 0.43, and 0.18, respectively. The proposed techniques strongly affirmed that for any patient to be discharged or die of the Covid-19, the patient must be confirmed Covid-19 positive, whereas the conventional method showed moderate to very weak affirmation. Based on the results, the proposed techniques are robust and show a very strong association between the variables of interest than the conventional techniques.

show abstract

Multivariate correlation analysis and its application in environmental analysis

Cited by 12 publications

References 2 publications

coastal aquifer by using correlation, principal component, and factor analyses

coastal aquifer by using correlation, principal component, and factor analyses

Multi-Variate Cross-Correlation and Image Matching.

Robust Multivariate Correlation Techniques: A Confirmation Analysis using Covid-19 Data Set

Contact Info

Product

Resources

About