Abstract
Variable analysis by means of ROC curves and categorical models. Formal relationship between the relative (or receiver) operating characteristics (ROC) models and the analytical models for categorical data is proposed. Traditionally, some differences have been established between signal detection theory (TDS) models and Generalized Linear Models (GLMs). However, some authors have suggested some specific relations (v.g. Dorfman y Alf, 1968; Swets, 1986; DeCarlo , 1998 and Tosteson y Begg, 1988). For example, the categorical models generate results in a contingence table similar to the conditional responses to TDS. Therefore, it is possible to include standard measures of association derived from statistical models (Bishop Fienberg and Holland, 1975).Hence, same measures are function of the cross-product ratio (independent of marginal totals) as LOR, η and Q . These indices are also consistent with a variable-criterion model (Swets, 1986, 1996), and ROC analysis can by applied. There are also other indices consistent whith ROC analysis that imply a threshold model. In short, we propound that it is possible to evaluate the empirical data by using two models that can be complementary: TDS and GLMs models.
