Discriminant function for classification of genuine and counterfeit naira notes
The aim of this work was to construct a model for the detection of counterfeit and genuine naira currency in Nigeria. Measurements based on three variables (mass, thickness and tensile stress) were obtained from twenty pieces of fake and genuine ₦1000 notes. A derivation of the appropriate measure of the separation of the two groups was made. Discriminant analysis technique was used to construct a discriminant function. Test for the significance of differences between the mean values of the variables for each of the two groups (counterfeit and genuine) of the ₦1000 notes was also carried out. Findings of the research revealed that the naira notes are to be classified as counterfeit or genuine according to the model: Z = 440.3007X - 858.8366Y + 147.5228Z such that Z > Z0 where Z0 is the end point of classification. Hotelling’s T2 and Mahalanobis quantity were also computed. The test result showed that differences between the counterfeit and genuine naira notes in the mean values of mass (X), thickness (Y) and tensile stress (Z) are significant.
Keywords: Mass, thickness, tensile stress, Hotelling’s T2 and Mahalanobis quantity