ملخص الدراسة:

In this research work an attempt has been made to find efficient estimations mean vector (location parameter) and variance - covariance matrix (scull parameter) for linear and quadratic discriminant function and to find the robust linear and quadratic discriminant functions when there is outliers by using a number of high quality robust methods estimation. of these methods are ( S – Estimators ) , M estimator by using weight ( Huber ) and M estimator by using weight ( Cambpe1l ) , moreover , comparing robust discriminant functions with classical discriminant functions to draw out the best discriminant function in order to diagnose two types of diagnosis diseases height rate cholesterol in blood which are : (Clot hearty and Clot stroke ) .