Resistant fits for some commonly used logistic models with medical application.

By Daryl Pregibon. 1982

How researchers see this paper

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

Citations in the corpus

Active Bias: Training More Accurate Neural Networks by Emphasizing High Variance Samples (2017, background): “On the other hand, to make models more robust to outliers, some approaches inject bias into the loss function in order to emphasize easier examples [37, 48, 27, 35].”
Comment: Demystifying Double Robustness: A Comparison of Alternative Strategies for Estimating a Population Mean from Incomplete Data. (2008, background): “Logistic models can be a poor way to estimate response propensities, because ML estimates from the logistic model are not resistant to outliers (Pregibon, 1982).”
Demystifying Double Robustness: A Comparison of Alternative Strategies for Estimating a Population Mean from Incomplete Data (2007, background): “Logistic models can be a poor way to estimate response propensities, because ML estimates from the logistic model are not resistant to outliers (Pregibon, 1982).”
A Wald-type test statistic for testing linear hypothesis in logistic regression models based on minimum density power divergence estimator (2016, background): “Hence, the effect of these (misclassification) error in response cannot be clearly inferred only from these influence functions; see Pregibon (1982), Copas (1988) and Victoria-Feser (2000) for more examples such analysis of misclassification error in logistic regression with a fixed design.”
Robust Logistic Regression and Classification (2014, background): “For example, Pregiobon [15] proposed the following M-estimator:”
Robit Regression: A Simple Robust Alternative to Logistic and Probit Regression (2005, methodology): “1 1 Introdu tionThe logisti and probit regression models are ommonly used in pra ti e to analyze binary responsedata, but many authors (see, Pregibon (1982) and the referen es therein) have shown that theirmaximum likelihood estimators are not robust.”
Model-based process monitoring using robust generalized linear models (2005, methodology): “It adopts the robust GLM procedure introduced by Pregibon (1982) and applies it to independent gamma-distributed responses.”
Robust estimation in generalized linear models: the density power divergence approach (2014, methodology): “The data set was obtained from a controlled study and analyzed by Pregibon (1982) and by Croux and Haesbroeck (2003).”
Robust Analysis of Generalized Linear Mixed Models (2004, methodology): “The nonrobustness properties of classical ML estimators in the framework of generalized linear models (GLM’s) have been studied by a number of authors (see, e.g., Pregibon 1982; Stefanski, Carroll, and Ruppert 1986; Kunsch, Stefanski, and Carroll 1989; Morgenthaler 1992; Carroll and Pederson 1993).”
Robust Learning from Untrusted Sources (2019, methodology): “, 2018) 14/71/0 179/331/0 390/120/0 432/78/0 472/38/0 502/8/0 503/7/0 497/13/0 (Pregibon, 1982) 55/30/0 308/202/0 361/149/0 416/94/0 437/73/0 455/55/0 470/40/0 485/25/0 Batch norm 0/85/0 107/403/0 317/193/0 416/94/0 446/63/1 478/32/0 487/23/0 482/28/0”
A General Method for Robust Bayesian Modeling (2015, methodology): “Pregibon (1982) and Stefanski et al. (1986) use M-estimators (Huber, 1964) to form more robust loss functions, which are designed to reduce the contribution from possible outliers.”
Robust Inference for Generalized Linear Models (2001, background): “The nonrobustness of the maximum likelihood estimator for ‚ has been studied extensively in the literature: see, for instance, the early work of Pregibon (1982) on logistic regression; Stefanski Carroll, and Ruppert (1986); Künsch, Stefanski, and Carroll (1989); Morgenthaler (1992); and Ruckstuhl…”