PROCEDURES FOR THE IDENTIFICATION OF MULTIPLE OUTLIERS IN LINEAR-MODELS
PROCEDURES FOR THE IDENTIFICATION OF MULTIPLE OUTLIERS IN LINEAR-MODELS
复制标题
DOI：
10.2307/2291266
复制
发表时间：
1993-12-01
影响因子：
3.7
通讯作者：
SIMONOFF, JS
中科院分区：
文献类型：
--
作者：
HADI, AS;SIMONOFF, JS
我们考虑在线性模型中识别和测试多个异常值的问题。可用的异常值识别方法通常无法成功检测多个异常值，因为它们受到应识别的观测值的影响。我们引入了两个测试程序来检测多个异常值，这些异常值似乎对此问题不太敏感。这两个过程都试图将数据分成一组“干净”的数据点和一组包含潜在异常值的点。然后使用预测误差的适当缩放版本来测试潜在的异常值，以了解它们相对于干净子集的极端程度。使用已知包含多个异常值的多个数据集来说明该过程并与各种现有方法进行比较。此外，这两个程序的性能都通过蒙特卡罗研究进行了调查。数据集和蒙特卡罗表明，这两种程序在检测线性模型中的多个异常值方面都是有效的，并且优于其他方法，包括基于稳健拟合的方法（例如，残差平方的最小中值）。特别是，这些方法不需要预设要测试的异常值数量，不需要估计器的效率水平，不需要蒙特卡罗来确定截止值，计算量不高，并且相对抵抗掩蔽和淹没效应。
We consider the problem of identifying and testing multiple outliers in linear models. The available outlier identification methods often do not succeed in detecting multiple outliers because they are affected by the observations they are supposed to identify. We introduce two test procedures for the detection of multiple outliers that appear to be less sensitive to this problem. Both procedures attempt to separate the data into a set of ''clean'' data points and a set of points that contain the potential outliers. The potential outliers are then tested to see how extreme they are relative to the clean subset, using an appropriately scaled version of the prediction error. The procedures are illustrated and compared to various existing methods, using several data sets known to contain multiple outliers. Also, the performances of both.procedures are investigated by a Monte Carlo study. The data sets and the Monte Carlo indicate that both procedures are effective in the detection of multiple outliers in linear models and are superior to other methods, including methods based on robust fits (e.g., least median of squares residuals). In particular, the methods do not require presetting numbers of outliers to test for, do not require the efficiency level of an estimator, do not require Monte Carlo to determine cutoff values, are not highly computationally intensive, and are relatively resistant to both masking and swamping effects.