189
mal distribution of data. This state of the obtained variables was improved by applying the transformation of the data set. For the normalization of a heterogeneous set
of values, the statistical method of Box-Cox is used (Box and Cox 1964). Normalized
values are processed with basic descriptive statistics, using the basic parameters of
this method.
Bivariate analysis is used to perform a comparative analysis of the dependencies
between variables (set of values for one variable). Two-dimensional scatter-plots are
used to display the correlations and identify them visually.
Multivariate analysis is applied to extract dominant associations of variables
(Šajn 2006). As a measure of similarity between variables, the product-moment correlation coefficient (r) was applied. There are various rotational strategies that have
been proposed (Šajn 2006; Žibret and Šajn 2010). The purpose of the applied statistical data processing models is to obtain the most realistic model of multi- dimensional
variable distribution system. For variables, i.e. the data set for the content of a given
element, for which the applied factor analysis will deposit low values, they will be
excluded from further analysis. Factor analysis used orthogonal varimax rotation.
Factor analysis is an interdependence technique because it looks for a group of variables that are similar in that they “move together” and therefore have great interdependence. When one variable has a large value, then the other variables in the group
have a large value. For the effective application of factor analysis, as well as other
multivariate interdependence techniques, it is necessary to have a minimal amount
of redundancy of variables, that is, the variables at least slightly overlap in their
meaning. Thanks to this redundancy, it is possible to discover a pattern in the behaviour of variables, that is, the basic idea (factor) by which they are imbued (Žibret and
Šajn 2010). Cluster analysis is a statistical technique used to identify how different
entities – variables – can be grouped together for the sake of their characteristics.
Also known as clustering, it is a preliminary data analysis tool that aims to sort different objects in a group in such a way that those who belong to the same group have
the highest degree of association. The most commonly processed dendrograms or
clusters are expressed by distance, Dlink/Dmax (%).
6.7.4 Evaluation of Data Summary Matrix
Data summary was introduced by Balabanova et al. (2019) generating the following
geochemical associations: F1, Ga-Nb-Ta-Y-(La-Gd)-(Eu-Lu); F2, Be-Cr-Li-Mg-Ni;
F3, Ag-Bi-Cd-Cu-In-Mn-Pb-Sb-Te-W-Zn; F4, Ba-Cs-Hf-Pd-Rb-Sr-Tl-Zr; F5,
As-Co-Ge-V; and F6, К-Na-Sc-Ti. The total variability for dominant loadings of
81.5% was established (Balabanova et al. 2019).
6 Evidence for Atmospheric Depositions Using Attic Dust, Spatial Mapping…
mal distribution of data. This state of the obtained variables was improved by applying the transformation of the data set. For the normalization of a heterogeneous set
of values, the statistical method of Box-Cox is used (Box and Cox 1964). Normalized
values are processed with basic descriptive statistics, using the basic parameters of
this method.
Bivariate analysis is used to perform a comparative analysis of the dependencies
between variables (set of values for one variable). Two-dimensional scatter-plots are
used to display the correlations and identify them visually.
Multivariate analysis is applied to extract dominant associations of variables
(Šajn 2006). As a measure of similarity between variables, the product-moment correlation coefficient (r) was applied. There are various rotational strategies that have
been proposed (Šajn 2006; Žibret and Šajn 2010). The purpose of the applied statistical data processing models is to obtain the most realistic model of multi- dimensional
variable distribution system. For variables, i.e. the data set for the content of a given
element, for which the applied factor analysis will deposit low values, they will be
excluded from further analysis. Factor analysis used orthogonal varimax rotation.
Factor analysis is an interdependence technique because it looks for a group of variables that are similar in that they “move together” and therefore have great interdependence. When one variable has a large value, then the other variables in the group
have a large value. For the effective application of factor analysis, as well as other
multivariate interdependence techniques, it is necessary to have a minimal amount
of redundancy of variables, that is, the variables at least slightly overlap in their
meaning. Thanks to this redundancy, it is possible to discover a pattern in the behaviour of variables, that is, the basic idea (factor) by which they are imbued (Žibret and
Šajn 2010). Cluster analysis is a statistical technique used to identify how different
entities – variables – can be grouped together for the sake of their characteristics.
Also known as clustering, it is a preliminary data analysis tool that aims to sort different objects in a group in such a way that those who belong to the same group have
the highest degree of association. The most commonly processed dendrograms or
clusters are expressed by distance, Dlink/Dmax (%).
6.7.4 Evaluation of Data Summary Matrix
Data summary was introduced by Balabanova et al. (2019) generating the following
geochemical associations: F1, Ga-Nb-Ta-Y-(La-Gd)-(Eu-Lu); F2, Be-Cr-Li-Mg-Ni;
F3, Ag-Bi-Cd-Cu-In-Mn-Pb-Sb-Te-W-Zn; F4, Ba-Cs-Hf-Pd-Rb-Sr-Tl-Zr; F5,
As-Co-Ge-V; and F6, К-Na-Sc-Ti. The total variability for dominant loadings of
81.5% was established (Balabanova et al. 2019).
6 Evidence for Atmospheric Depositions Using Attic Dust, Spatial Mapping…