We consider a population partitioned in several classes, a priori identified by a qualitative character with K levels; for a training sample TS of individuals the class of belonging is known, as well as the values of p explanatory characters observed on them. The descriptive aim of the discrimination is to check to what extent on TS the explanatory characters are able to distinguish among the classes and to provide a suitable explanation of the obtained distinction. The following decision aim of the discrimination is to use the previous results to assign further individuals of the population to a class, according only to the nown values taken by the $p$ explanatory characters on the new individuals. In this paper we limit our attention to the 2-classes case, when the explanatory characters are qualitative. The Scoring method aims at calculating a global score for every individual by summing up, for each individual, the scores corresponding to the observed levels of each explanatory character. The level's scores are those previously defined by the descriptive phase carried out on the training sample TS. Based on the obtained score, the individual is affected to either class or no decision is taken. The application proposed here uses the Spad software to show how the scores are attributed to the explanatory characters levels and then to check the quality of the method.
Scheda prodotto non validato
Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo