ALTERNATIVE TO RIDGE ESTIMATOR WITHOUT RIDGE PARAMETER
A .V. DORUGADE *
Y. C. Mahavidyalaya, Halkarni, Tal- Chandgad, Kolhapur (MS) - 416552, India.
*Author to whom correspondence should be addressed.
Abstract
The problem of estimation of the regression coefficients in a multiple regression model is considered under multicollinearity, heteroscedasticity and outlier situations. This paper introduces two estimators for regression parameters. The merit of the proposed estimators is that it does not require estimating the ridge parameter unlike other existing estimators. Numerical and simulated data sets are used to demonstrate the performance of the proposed estimators with other estimators in the sense of smaller MSE criteria. New estimators have desirable robustness features against the multicollinearity, heteroscedasticity and outlier situations; hence these are promising and can be recommended to the practitioners.
Keywords: Multicollinearity, ridge estimator, heteroscedasticity, outlier, mean square error, simulation