Two important problems are considered in regression analysis;
multicollinearity and the existence of influential data points. The ordinary least
squares estimators (LS) of coefficients are known to possess certain optimal proper-
120 Moawad El-Fallah Abd El-Salam
ties when explanatory variables are not correlated among themselves, and the
disturbances of the regression equation are independent, identically distributed
normal random variables. The presence of correlation among the explanatory
variables may result in imprecise information being available about the regression
coefficients. In addition, the least squares estimator may produce extremely poor
estimates in the presence of leverage or influential data points. Thus, various
remedial techniques have been suggested for these problems separately. One such
remedial technique is ridge regression to deal with multicollinearity, and the
robust estimation techniques are not as strongly affected by the presence of
influential data points. However, although, we usually think of these two
problems separately, but in practical situations, these problems occur
simultaneously. To remedy these two problems simultaneously, several robust
ridge regression estimators have been put forward that are much less influenced
by the influential data points and multicollinearity. Askin and Montgomery
(1980), suggested combining the ridge and the least absolute deviation (LAD)
robust regression techniques. Montgomery and peck (1982), have suggested that
either robust or ridge estimation methods alone may be sufficient for dealing with
the combined problem. In this paper, we take the initiative to develop a more
robust technique to remedy these two problems. We proposed combining the ridge
regression with the highly efficient and high breakdown point estimator, namely
the Ridge Least Median Squares (RLMS) estimator. We call this modified
method, the robust ridge regression based on Least Median Squares estimation
(RLMS). We expect that, the modified method would be less sensitive to the
presence of influential points and multicollinearity. So, the aim of this paper is
devoted to examine some estimators which are resistant to the combined problems
of multicollinearity and influential points. Exactly, can the ridge estimators and
some robust estimation techniques be combined to produce a robust ridge
regression estimator?. The remainder of the paper is organized as follows. In
section (2), the ridge regression estimator will be reviewed. The robust regression
estimation will be discussed in section (3). In section (4), we discuss the
augmented ridge robust estimators as a way of combining biased and robust
regression techniques, while, Section (5) introduces the proposed combined ridge
robust estimator (RLMS). Section (6) presents the results of a Monte Carlo
simulation study to investigate how such estimators perform well, and some
concluding remarks are presented in section (7).