27
1.5.1 Correlations and Similarity Models: Challenges
and Atypical Implementations
The correlation is interdependence and mutual dependence, i.e., mutual connection
and dependence, the relationship between two economic phenomena that will be
found in causal dependence and conditioning. The correlation can be parallel, i.e.,
positive and, vice versa, negative. The correlation is positive when the growth of one
phenomenon causes the growth of the other (e.g., the growth of the demand causes
the growth of the price) or the reduction of one phenomenon causes the reduction of
the other. The correlation is negative when the growth of one phenomenon is followed by the fall of the other (e.g., the ratio of changes in price and demand) or,
conversely, the reduction of one phenomenon is followed by growth of the other.
The intensity of the correlation is measured by the correlation coefficient. The
notion of correlation in many ways coincides with the notion of function.
Bivariate analysis is used to perform a comparative analysis of the dependencies
between variables (set of values for one variable). Two-dimensional scatterplots 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 (Žibret and Šajn 2010; Šajn 2006).
The purpose of the applied statistical data processing models is to obtain the
most realistic model of multi-dimensional variables’ 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.
1.5.2 Anomalous and Atypical Occurrence
Newer laboratory instruments, i.e., their software, usually write the data in binary
files, in accordance with the so-called good laboratory practice. Data written in this
way can be more difficult to manipulate, and this is especially important when analyses have potential legal consequences. However, if we put aside such illegitimate
manipulations, users occasionally have some requirements that the software of a
given instrument did not anticipate or allow. Very often the graphical display of the
experimental data is not as required by the user. Sometimes it is also necessary to
perform a more detailed analysis or decomposition of peaks (deconvolution). There
are a number of statistical packages and tools that enable the processing of raw data
obtained from analytical instruments.
In most cases, the data obtained from the instrumental measurements, in addition
to the measured signal, also contain certain noises due to certain influences on the
measuring instrument. One of the possibilities is to “filter” the obtained data with
program tools contained in the software of the analytical tool. Several noise filtering
1 General Aspects of Environmental Degradation vs. Technological Development…
1.5.1 Correlations and Similarity Models: Challenges
and Atypical Implementations
The correlation is interdependence and mutual dependence, i.e., mutual connection
and dependence, the relationship between two economic phenomena that will be
found in causal dependence and conditioning. The correlation can be parallel, i.e.,
positive and, vice versa, negative. The correlation is positive when the growth of one
phenomenon causes the growth of the other (e.g., the growth of the demand causes
the growth of the price) or the reduction of one phenomenon causes the reduction of
the other. The correlation is negative when the growth of one phenomenon is followed by the fall of the other (e.g., the ratio of changes in price and demand) or,
conversely, the reduction of one phenomenon is followed by growth of the other.
The intensity of the correlation is measured by the correlation coefficient. The
notion of correlation in many ways coincides with the notion of function.
Bivariate analysis is used to perform a comparative analysis of the dependencies
between variables (set of values for one variable). Two-dimensional scatterplots 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 (Žibret and Šajn 2010; Šajn 2006).
The purpose of the applied statistical data processing models is to obtain the
most realistic model of multi-dimensional variables’ 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.
1.5.2 Anomalous and Atypical Occurrence
Newer laboratory instruments, i.e., their software, usually write the data in binary
files, in accordance with the so-called good laboratory practice. Data written in this
way can be more difficult to manipulate, and this is especially important when analyses have potential legal consequences. However, if we put aside such illegitimate
manipulations, users occasionally have some requirements that the software of a
given instrument did not anticipate or allow. Very often the graphical display of the
experimental data is not as required by the user. Sometimes it is also necessary to
perform a more detailed analysis or decomposition of peaks (deconvolution). There
are a number of statistical packages and tools that enable the processing of raw data
obtained from analytical instruments.
In most cases, the data obtained from the instrumental measurements, in addition
to the measured signal, also contain certain noises due to certain influences on the
measuring instrument. One of the possibilities is to “filter” the obtained data with
program tools contained in the software of the analytical tool. Several noise filtering
1 General Aspects of Environmental Degradation vs. Technological Development…