METHODS FOR MULTIVARIATE DATA ANALYSIS IN THE STUDY OF ORAL DISEASES: THE MULTIPLE LINEAR REGRESSION (Abstract): During the statistical analysis of medical data, in many situations it is necessary to identify the multiple correlations
established between the studied parameters. In this purpose, one of the most useful methods is to build a model of multiple regression, which allows the modeling of a dependant variable values having at least the ordinal type, based on its linear relation with more than one independent variables satisfying the same restriction, called predictors. The multiple linear regression model is
a generalization of the simple linear regression model, which identifies the parameters of an equation with n variables y = b0+b1x1+b2x2+…+ bnxn+e, based on which we can find the predictors that have a statistically significant influence over the dependant variable. We used this model to identify, on a set of 202 patients having different types of oral lesions, the biochemical
analysis which can be eventually correlated with the oral diagnosis. We found that the values of leucocytes, hemoglobin and hematocrit are significant for the general oral diagnosis, the cholesterol and glucose values for the oral lesion type, and the hemoglobin for the periodontal disease. The identified predictors are useful for further data processing.