With the help of the valid and approved model, we can substitute values different from the original ones, which allows us to study processes that currently cannot be measured in reality but may occur by certain probability. We can thus obtain relevant and new information with the help of which we can establish our future decisions.
During modelling, it is always to be examined to what extent simulation results can be considered realistic responses of the modelled system. The regular solution in the practice of modelling is to compare simulation results with the values measured on the modelled attribute in the system we want to describe. Simulation can be considered ideally proper and accurate if the measured values and simulation results calculated on the same attribute are the same.
With the use of the model with different input values, we can examine several outputs by which we can determine the most likely scenarios.
It shows the strength and direction of the relation between two criteria or variables. Its practical use is that during the analysis of a particular element, we can find aspects in the mass of facts that shows similar features to or different from the one we want to examine. For example, the person who has a higher income will spend more on clothing than the person with lower income. It means that there is a positive correlation between the two criteria or variables.
The correlation shows only the strength and direction of the relation, whereas regression defines function, so substituting any values we can specify the result of the target value. Using the example we used with correlation, the regression function shows how high expenditure on clothes we can expect with a particular income level.
After data structuring and using the proper statistical method (cluster analysis), we can explore typical patterns, with the help of which we can define groups specified by particular criteria.