Robust ridge estimation methods for predicting u. s. coal mining fatalities

This paper compares alternative combinations of ridge regression and robust regression techniques in predicting fatalities in the U.S. coal mining industry. The combined robust ridge approach is needed because of the simultaneous occurrence of multicollinearity and outliers in the data. Not surprisingly, the relative effectiveness of different ridge estimators in predicting coal mining fatalities is found to be dependent upon the particular choice of the robust weighting scheme employed.