Hossein Shahi - Department of mining engineering, university of Gonabad
Reza Ghavami - Faculty of mining, Petroleum and Geophysics, Shahrood university of technology

Discriminant function is a multivariate statisticalmethod that has been used to determine correlationbetween every sample with one of known societiesbased on their litho geochemical analysis. In thisstudy, geochemical features of elements that areprobably related to mineralization of Au and Cudeposit in Tanurcheh area ,east of Iran, wereevaluated using discriminant function model. Inorder to assessing of correct evaluation, discriminantfunction was utilized over Au element. Discriminantfunction method demonstrates that there is not highrelationship between the elements of Au and Cu andthere are probably two mineralizing types of Cuporphyry and veiny Au in this district. In the finallythe variety of elements that are related to anomalyand background in veiny Au mineralization typewere identified and after classifying the samples, theability and validity of discriminant function methodwere discussed. Leave-one-out cross- validation as apopular strategy was utilized to estimate the risk ofan estimator. With regard to obtained functions, thenew samples in the region can be classified easilyinto anomaly, background and transition subpopulationsfor Au mineralization.

Discriminant function model, veiny Au deposit, classification, Leave-one-out crossvalidation