Local maximum likelihood estimation and inference

By Jianqing Fan, Mark W. Farmen, Irène Gijbels. 1998

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

Generalized random forests (2016, methodology): “A popular approach to estimating such functions θ(x) is to first define some kind of similarity weights αi(x) that measure the relevance of the i-th training example to fitting θ(·) at x, and then fit the target of interest via an empirical version of the estimating equation (Fan et al., 1998; Newey”
Iterative Weighted Maximum Likelihood Denoising With Probabilistic Patch-Based Weights (2009, background): “the variance of the estimate at the cost of a bias introduced by samples that follow a distribution with a parameter different to [31].”
A General Framework for Estimation and Inference of Geographically Weighted Regression Models: 1. Location-Specific Kernel Bandwidths and a Test for Locational Heterogeneity (2002, background): “Although some issues pertaining inference of local models are starting to be addressed in the smoothing literature (see Fan et al, 1998), it is not clear that these developments will carry over placidly to the spatial case, thus making the attempt of developing GWR into a formal model a difficult…”
Statistical Methods with Varying Coefficient Models. (2008, methodology): “Fan et al. (2003) employed the local constant approach to estimate the coefficients α0(t) and α1(t) in a similar manner to Section 2.”
Statistical Estimation in Generalized Multiparameter Likelihood Models (2009, methodology): “For local likelihood estimation, Aerts and Claeskens (1997) considered cross-validation and plug-in bandwidths, and Fan, Farmen, and Gijbels (1998) suggested a bandwidth selector based on an approximation to the integrated mean squared error.”
Non-Parametric Regression for Binary Dependent Variables (2006, methodology): “Locally dropping regressors that cause collinearity reduces the 18 Fan et al. (1998) suggested to use only bandwidth values that are sufficiently large such that all local smoothing windows contain zeros as well as ones.”
Oriented Object Proposals (2015, background): “(1) can be achieved by maximizing the likelihood of a set of neighbor pixels [13] [23] where l n is the likelihood of a pixel belonging to an object. x n is the pixel position.”
Local Likelihood Estimation in Generalized Additive Models (2003, methodology): “Using this estimator as a starting point, Linton (2000) proposed a local likelihood procedure to find a fully efficient estimator for individual additive component functions when the remaining component functions are suitably undersmoothed (see also Fan et al., 1998b).”
Local likelihood smoothing of sample extremes (2000, methodology): “In applications of kernel smoothing the choice of kernel is not usually critical from a practical point of view, and in linear smoothers there are theoretical grounds for preferring the Epanechnikov kernel, which is optimal in a certain sense (Fan and Gijbels (1996), page 75).”
Tilted Nonparametric Estimation of Volatility Functions With Empirical Applications (2010, unknown): “maximum likelihood (LML) estimators, given by (3), (11) and ( 13 ) respectively, along with”