Abstract.Regression model building is often used in studying the relationship between a dependent variables and a set of independent variables. The effects of outliers on linear regression are examined. To improve this problem, we introduce a new statistical techniques mean median regression have been developed that are not so easily affected by outliers. This is the robust (or resistant) method. The performance of mean median regression estimates is compared with absolute deviation estimates in a simulation study, and we discuss the test for correct specification of the disturbances mainly because of their ability to detect also irregularities in the regressor specification.

Received: March 31, 2007

AMS Subject Classification: 62J02

Key Words and Phrases: efficient estimator, regression median, goal programming, simulation study

Source: International Journal of Pure and Applied Mathematics
ISSN: 1311-8080
Year: 2007
Volume: 37
Issue: 4