An Empirical Investigation of the Behavior of Bounded-influence Regression Estimators |
Contents
THE HISTORY OF ROBUST ESTIMATES OF LOCATION | 9 |
THE FAILINGS OF ORDINARY LEAST SQUARES REGRESSION | 24 |
THE DEVELOPMENT OF ROBUST REGRESSION | 38 |
5 other sections not shown
Common terms and phrases
asymptotic variance behavior BIF estimators BIF Regressions Relative BIF solution identical BIFMOD bisquare bounded-influence estimators Chapter choice coefficients and scale collinearity Combined Total Fraction Combined Total Leverage computed Controlled Data Points controlled point covariance matrix cut-off values data sets Default Sensitivity Bounds DFFIT downweight e/SSE Fraction of Residual fraction of SSE Gaussian gross errors Hampel high leverage points Huber estimator indicates a BIF influence curve influence function influential subsets initial estimate key point Krasker and Welsch Krasker-Welsch estimator leave-out-one solution leverage and residual Leverage for Fraction linear M-estimator Mallows algorithm Mallows estimator normal distribution outliers parameters Performance of BIF principal component procedures redescending regression M-estimate relative efficiency Relative to OLS residual combination Residual Sum robust estimators robust regression sample mean scale estimate Schweppe estimator Section singular value decomposition singular values Specified at 95 Sum of Squares Tabled values variance-covariance matrix vector Velleman and Ypelaar weight function