Detection of Influential Observation in Linear Regression

By R. Dennis Cook. 2000

3,246 citations recorded by the source corpus.

Subfields: Applied Mathematics, Management Science and Operations Research, Statistics and Probability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Applied Mathematics

In Management Science and Operations Research

In Statistics and Probability

Citations in the corpus

Explainable machine learning in deployment (2019, methodology): “This technique asks the question: Which data point in the training dataset x ∈ Dx is most influential to the model’s output f (x test) for a test point x test? Statisticians have used measures like Cook’s distance [15] which measure the effect of deleting a data point on the model output.”
DeltaGrad: Rapid retraining of machine learning models (2020, methodology): “Our method can be also used to evaluate the importance of training samples (see Cook (1977) and the follow-up works such as Ghorbani & Zou (2019)).”
BART: Bayesian Additive Regression Trees (2008, methodology): “diagnostic Dx for each x ( Cook 1977 ) based on a linear least squares regression”
Data Shapley: Equitable Valuation of Data for Machine Learning (2019, methodology): “M L ] 5 A p r 2019 of leave-one-out (LOO) test: compare the difference in the predictor’s performance when trained on the full dataset vs. the performance when trained on the full set minus one point (Cook, 1977).”
Jeopardy, Non-Public Information, and Insider Trading Around SEC 10-K and 10-Q Filings (2005, neutral): “gression results we eliminate outliers using Cook’s (1977) distance statistic.”
The dualistic regulatory effect of passion on the relationship between fear of failure and negative affect: Insights from facial expression analysis (2020, methodology): “We tested the hypotheses using robust regression (Cook, 1977; Hamilton, 1991), rreg in Stata 14, as it resists the pull of outliers and generates robust standard errors.”
Foreign direct investment and technology spillover: A cross-industry analysis of Thai manufacturing (2006, background): “Cook’s distance is the ‘influence statistic’ developed by Cook (1977).”
Applied Linear Regression (2004, methodology): “3 shows a residual plot for the fit of the mean function E(Y|X = x) = β0 + β1x1 + β2x2 for the artificial data given in the file caution from Cook and Weisberg (1999b). The right-opening megaphone is clear in this graph, suggesting nonconstant variance.”
Linear Regression Diagnostics (2015, neutral): “) 1 p 1-h1 A measure similar to this has been suggested by Cook [14].”
Losing Sight of the Forest for the Trees? Productive Capabilities and Gains From Trade as Drivers of Vertical Scope (2005, methodology): “Second, we corroborate this analysis using robust regressions7 ( Cook, 1977;”
The Shapley Value in Machine Learning (2022, methodology): “The Shapley value is not the only method for data valuation – earlier works used function utilization [Koh et al. , 2017; Sharchilev et al. , 2018], leave-one-out testing [Cook, 1977] and core sets [Dasgupta et al. , 2009].”
Do institutional investors exploit the post-earnings announcement drift? (2005, positive): “2822 outliers are deleted using Cook’s (1977) distance statistic.”
DataComp: In search of the next generation of multimodal datasets (2023, methodology): “Such methods include influence functions, a classic technique from robust statistics [52, 31] that uses a second-order Taylor expansion to approximate the effect of removing a training point on the learned model parameters [74, 75, 53, 48], as well as methods that fit attribution functions directly ”