Maximum Penalized Likelihood Estimation

By Bernard Walter Silverman. 2006

135 citations recorded by the source corpus.

Subfields: Applied Mathematics, Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/13a29b02f4d831bb42ee91eab5d49eff0a7eebda

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 Artificial Intelligence

Citations in the corpus

Estimation of nonparametric conditional moment models with possibly nonsmooth moments (2008, background): “As already mentioned in Section 2, from the well-known 18 results on convex optimization (see, e.g., Eggermont and LaRiccia, 2001), when E [m(X,h)′m(X,h)] and P (h) are convex, and H is closed and convex, the constraint optimization problem h0 ∈ M0 ≡ { h : arg inf h∈H”
A kernel density estimation method for networks, its computational method and a GIS‐based tool (2009, methodology): “Many methods for treating the edge problem have been proposed in the literature of the univariate kernel method, and those methods can be applied to )1( L (see among others Devroye and Gyorfi (1985); Eggermont and LaRiccia (2001); Scott (1992); Silverman (1986); Tapia and Thompson (1978)).”
ADAPTIVE SEMI-VARYING COEFFICIENT MODEL SELECTION (2012, background): “, n, are quasi-uniform (Eggermont and LaRiccia (2009)), one can verify that as long as λ0nh −1/2 → 0 but λ0n 3/2 → ∞, the two conditions listed in (2.”
Estimation of Nonparametric Conditional Moment Models with Possibly Nonsmooth Generalized Residuals (2011, methodology): “Then the PSMD procedure (6) is equivalent to Q n ( h n ) + ν n R n ( h n ) ≤ inf h∈clsp(Hn) Q n (h) + ν n R n (h) + O p (η n ), (10) where R n ( h n ) ≤ B n and ν n ≥ 0 is such that ν n (R n ( h n ) − B n ) = 0; see Eggermont and LaRiccia (2001).”
A smoothed backbone-dependent rotamer library for proteins derived from adaptive kernel density estimates and regressions (2011, unknown): “This is a common procedure in density estimations (Eggermont and LaRiccia, 2001).”
Choice between Semi‐parametric Estimators of Markov and Non‐Markov Multi‐state Models from Coarsened Observations (2007, background): “If no parametric assumptions are made about the functions to be estimated and if smooth estimators are favoured, the two main approaches are sieve estimators, extending the so-called hazard regression of Kooperberg & Clarkson (1997) or using orthogonal expansions such as in Müller & Stadtmüller (2”
Maximum likelihood estimation of a multi‐dimensional log‐concave density (2008, methodology): “, 1996), splines (Eubank, 1988; Wahba, 1990), penalized likelihood (Eggermont and LaRiccia, 2001) and vector support methods (Vapnik and Mukherjee, 2000).”
Model selection via testing: an alternative to (penalized) maximum likelihood estimators (2006, background): “leads to similar technical difficulties, as can be seen from [5,21,22,59] or [31].”
On the asymptotics of penalized spline smoothing (2011, methodology): “Specifically, Eggermont and LaRiccia [4, 6] have shown that, for any t ∈ [0, 1], φ̂(t) = ∫ 1 0 Kλ(t, s)f(s)ds+ 1 n n∑ i=1 Kλ(t, ti)ǫi + higher order terms, (15) where the equivalent kernel Kλ(t, s) is the corresponding Green’s function for the following ordinary differential equation with boundary c”
MAXIMUM SMOOTHED LIKELIHOOD ESTIMATION AND SMOOTHED MAXIMUM LIKELIHOOD ESTIMATION IN THE CURRENT STATUS MODEL (2010, methodology): “The first estimator we study is a so-called maximum smoothed likelihood estimator (MSLE) as studied by Eggermont and LaRiccia (2001) in the context of monotone and unimodal density estimation.”