Distance squared method fortified (QDE) and compared with some other estimation methods fortified with practical application


The Logistic Regression Model regarded as the important models which used in the classification, particularly in medical and biological fields. As the ordinary regression model, the logistic model be affected by the strange observations and its estimators will be sensitively influenced by the outliers, consequently, these estimators will be inaccurate and infeasible. Therefore, the researchers’ interesting focused on the “robust methods” as new suggestions toward number of problems or functions. Although, there are several approaches deal with this subject, but they still aim to use some approach that can make balancing among the observations by applying weights associated with those observations which thought to be outliers, and with smaller weight from those ones which associated with the rest data. Thus, this paper will be focused for a recent robust method called “Quadratic Distance Estimator IQ.D.E.1”. Moreover, Some other methods like: R & M achieved. The most important goal in this paper is to study and compare some of the robust methods and with commutative presence . Also using Quadratic Distance Estimator method on real data of patients which taking from Babylon hospital for the purpose of estimate parameters of binary logistic regression model