论文信息 - A Multistage Procedure for Detecting Several Outliers in Linear Regression

A Multistage Procedure for Detecting Several Outliers in Linear Regression

A new statistic, Fk , is proposed for detecting multiple outliers in linear regression. This statistic is incorporated into the following multistage procedure: Initially, a subset of k observations is selected to be tested. If Fk is found to be significant, the most extreme observation in the subset as determined by the largest studentized residual is deleted and the test repeated for the (k – 1) observations in the subset using the remaining sample. The procedure is stopped when a test fails to reject the no-outlier hypothesis. A Monte Carlo study is used to evaluate the performance of this procedure.

Mervyn G. Marasinghe

[1] J. W. Gorman,et al. Fitting Equations to Data. , 1973 .

[2] R. H. Moore,et al. Testing for a Single Outlier in Simple Linear Regression , 1973 .

[3] R. H. Moore,et al. Some Grubbs-Type Statistics for the Detection of Several Outliers , 1972 .

[4] Richard E. Lund,et al. Tables for An Approximate Test for Outliers in Linear Models , 1975 .

[5] Bernard Rosner,et al. On the Detection of Many Outliers , 1975 .

[6] J. F. Gentleman,et al. Detecting Outliers. II. Supplementing the Direct Analysis of Residuals , 1975 .

[7] N. Draper,et al. On Testing for Two Outliers or One Outlier in Two-Way Tables , 1978 .

[8] Douglas M. Hawkins,et al. Fractiles of an extended multiple outlier test , 1979 .

[9] Douglas M. Hawkins. Identification of Outliers , 1980, Monographs on Applied Probability and Statistics.

[10] M. R. Mickey,et al. Note on the use of stepwise regression in detecting outliers. , 1967, Computers and biomedical research, an international journal.