A New Approach to Identify Recharge Areas in the Lower Virgin River Basin and Surrounding Basins by Multivariate Statistics

Asante, Joseph; Kreamer, David K.

doi:10.1007/s11004-015-9583-0

Cited by 10 publications

(7 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The analysis was conducted in SPSS, which forms clusters of objects according to the squared Euclidean distances among those objects. Cluster analysis is commonly used to study groupings in other disciplines, such as geoscience (Asante & Kreamer, 2015) and medicine (Sonğur & Top, 2016). However, the use of cluster analysis is a novel approach to analyze ETR.…”

Section: Methodsmentioning

confidence: 99%

Offshoring in the European Union: a Study of the Evolution of the Tax Burden

Contell

Serrano

Serer

2017

View full text Add to dashboard Cite

Section: Methodsmentioning

confidence: 99%

Offshoring in the European Union: a Study of the Evolution of the Tax Burden

Contell

Serrano

Serer

2017

View full text Add to dashboard Cite

“…Multivariate clustering (MC) utilises unsupervised machine learning techniques to detect natural data clusters (Asante & Kreamer, 2015;Pászto et al, 2020). MC is a toolset that can be used to identify and map the locations of statistically significant spatial outliers (Peeters et al, 2015).…”

Section: Impact On Covid-19 Prevalence Of Demographic and Socioeconomic Variablesmentioning

confidence: 99%

Demographic and socioeconomic determinants of COVID-19 across Oman - A geospatial modelling approach

et al. 2021

View full text Add to dashboard Cite

Local, bivariate relationships between coronavirus 2019 (COVID-19) infection rates and a set of demographic and socioeconomic variables were explored at the district level in Oman. To limit multicollinearity a principal component analysis was conducted, the results of which showed that three components together could explain 65% of the total variance that were therefore subjected to further study. Comparison of a generalized linear model (GLM) and geographically weighted regression (GWR) indicated an improvement in model performance using GWR (goodness of fit=93%) compared to GLM (goodness of fit=86%). The local coefficient of determination (R2) showed a significant influence of specific demographic and socioeconomic factors on COVID-19, including percentages of Omani and non-Omani population at various age levels; spatial interaction; population density; number of hospital beds; total number of households; purchasing power; and purchasing power per km2. No direct correlation between COVID- 19 rates and health facilities distribution or tobacco usage. This study suggests that Poisson regression using GWR and GLM can address unobserved spatial non-stationary relationships. Findings of this study can promote current understanding of the demographic and socioeconomic variables impacting the spatial patterns of COVID-19 in Oman, allowing local and national authorities to adopt more appropriate strategies to cope with this pandemic in the future and also to allocate more effective prevention resources.

show abstract

“…BoxCox transformations (BOX & COX, 1964) of all single variables do not guarantee symmetry of the distribution, but more closeness to them (ASANTE & KREAMER, 2015;TEMPL et al, 2006). The applied transformation is a modification of the power transformation by BOX & COX (1964).…”

Section: Data Pre-processingmentioning

confidence: 99%

“…The study focused on trying to determine the right number of clusters and to analyse some suggested sum of squares indexes (called WB indexes). The "leave-one out" (LOO) classi fication method was used in the discriminant function analysis (DFA) as cross validation (ASANTE & KREAMER, 2015).…”

Section: Selecting the Proper Number Of Cluster Solutionsmentioning

confidence: 99%

“…To select the optimal number of clusters in the final solution, a statistics test based on the sum of squares was applied. Since a single statistics test method cannot be depended upon, additional methods were used (ASANTE & KREAMER, 2015). There are several suggested indexes depending on the sum of squares (Eq.2-5): Hartigan (1975…”

Section: Selecting the Proper Number Of Cluster Solutionsmentioning

confidence: 99%

See 1 more Smart Citation