Generalized Additive Models

By Trevor J. Hastie, Robert Tibshirani. 1986

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

Deep learning in neural networks: An overview (2014, neutral): “ri (Nowlan and Hinton, 1992). Other weight priors are implicit in additional penalty terms (MacKay, 1992) or in methods based on validation sets (Mosteller and Tukey, 1968; Stone, 1974; Eubank, 1988; Hastie and Tibshirani, 1990; Craven and Wahba, 1979; Golub et al., 1979), ﬁnal prediction error (Aka”
Hierarchical Mixtures of Experts and the EM Algorithm (1993, neutral): “IRLS also (cid:12)gures prominently in a branch of nonparametric statistics known as generalized additive models (GAM's; Hastie & Tibshirani, 1990).”
Demystifying Double Robustness: A Comparison of Alternative Strategies for Estimating a Population Mean from Incomplete Data (2007, neutral): “To check the appropriateness of the link function, Hinkley (1985) suggested adding η̂2 i as another covariate to see whether it is related to the response; the coefficient for this extra term was not significantly different from zero (p = 0.”
Sparse additive models (2007, positive): “Data version of sparse additive models To obtain a sample version of the population solution, we insert sample estimates into the population algorithm, as in standard backfitting (Hastie and Tibshirani, 1999).”
Component selection and smoothing in multivariate nonparametric regression (2006, positive): “The SS-ANOVA generalizes the popular additive model by Hastie and Tibshirani (1990) and provides a general framework for nonparametric multivariate function estimation.”
Sparse canonical correlation analysis (2009, unknown): “Theorem 1 is well known in the statistics community and corresponds to the equivalence between one form of Alternating Conditional Expectation (ACE) and CCA (Breiman & Friedman, 1985; Hastie & Tibshirani, 1990).”
Association of Sleep-Disordered Breathing, Sleep Apnea, and Hypertension in a Large Community-Based Study (2000, unknown): “To assess the impact of using arbitrary cutoff points, the OR of hypertension in relationship to SDB variables was also estimated using nonparametric logistic regression.(36,37) With this method, the dose-response relationship between a putative risk factor (eg, AHI) and an outcome (eg, hypertension”
Bayesian measures of model complexity and fit (2002, unknown): “…trace of the hat matrix is a standard result in linear modelling and has been applied to smoothing (Wahba, 1990) (page 63) and generalized additive models (Hastie and Tibshirani (1990), section 3.5), and is also the conclusion of Hodges and Sargent (2001) in the context of general linear models.”
Greedy function approximation: A gradient boosting machine. (2001, unknown): “Separate line searches in each such \subset" corresponds to applying (13) separately to the data in each terminal node of the decision tree lm = medianW yi Fm 1(xi) h(xi; am) xi 2 Rlm ; wi = jh(xi; am)j; (14)”
Prospective sampling based on model ensembles improves the detection of rare species (2010, unknown): “For each species we: (1) prepared species and environmental data, (2) modeled habitat suitability using two different techniques generalized additive model (GAM; Hastie and Tibshirani 1986), requiring presence/absence data, and ecological niche factor analysis (ENFA, Hirzel et al.”
Generalized additive models for large data sets (2015, unknown): “Current fitting methods for GAMs and related models (e.g. Wood (2011)) are reasonably efficient and robust when applied to data sets containing up to a few tens of thousands of observations, but they tend to become too memory intensive much beyond this point, so larger data sets, containing hundreds”
Actuarial Theory for Dependent Risks: Measures, Orders and Models (2005, unknown): “A standard reference for GAMs is Hastie and Tibshirani (1990). The basis of GAMs is the assumption that the data are sampled from a one-parameter exponential family of distributions with log-likelihood of the form (7.”
NONADDITIVE EFFECTS OF THE ENVIRONMENT ON THE SURVIVAL OF A LARGE MARINE FISH POPULATION (2004, unknown): “In the continuous case, Model 1 is enriched by the inclusion of second order interaction terms, whose effect is regulated by the nonparametric functions s (Hastie and Tibshirani 1990).”