The Use of the Discriminant Analysis for Obtaining Audit Evidence (II)
Author:
Univ. Prof. PhD Elisabeta JABA, Ioan-Bogdan ROBU
JEL:
DOI:
Keywords:
discriminatory analysis, influence factors, score function, regional map
Abstract:
The hypothesis that sights the determination of an entity’s situation at a certain time, during an audit engagement, having the discriminatory analysis as work method, have been ratified by the application of the SPSS15.0 statistic tool. A certain number of outcomes appeared based on the information in the data base that has been created; the outcomes will constitute the fundament for shaping and enabling the calculated discriminatory functions. The methodological approach proposed by the statistic tool targets the testing of the independent variables (means, variations, tolerance), also targets the impact of the independent variables over the discrimination and over the proposed mathematic models with the assistance of some statistics, the correlations at the discriminating functions level, but also the accuracy of the canonic coefficients resulted from the model.\r\nThe outputs generated by the program enable the discrimination of the main factors that might influence the situation of a firm at a certain time, the relationships and inter correlations between these and the manifestation intensity of these relationships. The calculated discriminating functions, combined following the creation of some canonic coefficients, enable the issuance of an opinion regarding the situation of a firm at a certain time, but also discussions about the influent factors.\r\nConsidering the hypothesis regarding the erroneous framing of a firm or regarding certain framings of new firms that are included, at a medium level, within the range of the quantified factors, there were created mathematic models for the forecast of framing a firm within a certain category. Through the discriminatory maps it was drown a classification and a graphic delimitation of the analyzed firms, on the three performance levels (leader, middle, least), taking into account the independent variables that participate at the score function shaping.\r\nThis method may constitute the premise for the thoroughness of the statistics influence within the theory and the practice of the financial audit; the statistical techniques opening new research fields and optimizing the current methods and practices.