Asymptotics: particles, processes and inverse problems. by Eric A. Cator, Cor Kraaikamp, Hendrik P. Lopuhaa, Jon A.

By Eric A. Cator, Cor Kraaikamp, Hendrik P. Lopuhaa, Jon A. Wellner, Geurt Jongbloed

Show description

Read Online or Download Asymptotics: particles, processes and inverse problems. Festschrift for Piet Groeneboom PDF

Best probability books

Probability and Theory of Errors (Fourth Edition)

This can be a pre-1923 historic copy that was once curated for caliber. caliber coverage used to be performed on each one of those books in an try and get rid of books with imperfections brought through the digitization strategy. although we now have made most sensible efforts - the books can have occasional blunders that don't abate the interpreting adventure.

Quantum Probability and Related Topics: Proceedings of the 30th Conference

This quantity comprises present paintings on the frontiers of analysis in quantum likelihood, limitless dimensional stochastic research, quantum info and information. It provides a delicately selected number of articles by way of specialists to spotlight the most recent d

Extra resources for Asymptotics: particles, processes and inverse problems. Festschrift for Piet Groeneboom

Example text

1, the fact that s1 belongs to L∞ is essentially a necessary condition to have a control on the L2 -risk of estimators of s1 . d. observations and the estimation of the intensity s = ns1 from a Poisson process. This suggests to adapt the known results from density estimation to intensity estimation for Poisson processes. We shall briefly explain how it works, when the Poisson process X has an intensity s ∈ L∞ (λ) with L∞ -norm s ∞ . The starting point is to observe that, given an element ϕ ∈ L2 (λ), a natural N estimator of ϕ, s is ϕ(X) = ϕ dΛX = i=1 ϕ(Xi ).

1, the fact that s1 belongs to L∞ is essentially a necessary condition to have a control on the L2 -risk of estimators of s1 . d. observations and the estimation of the intensity s = ns1 from a Poisson process. This suggests to adapt the known results from density estimation to intensity estimation for Poisson processes. We shall briefly explain how it works, when the Poisson process X has an intensity s ∈ L∞ (λ) with L∞ -norm s ∞ . The starting point is to observe that, given an element ϕ ∈ L2 (λ), a natural N estimator of ϕ, s is ϕ(X) = ϕ dΛX = i=1 ϕ(Xi ).

Let Φ(x) = sup{h(x) : h is convex and h(z) ≤ Φ(z) for all z ∈ [0, 1]}. Then sup |Φ(x) − Ψ(x)| ≤ sup |Φ(x) − Ψ(x)|. 0≤x≤1 0≤x≤1 Proof. Note that for all y ∈ [0, 1], either Φ(y) = Φ(y), or y is an interior point of a closed interval I over which Φ is linear. For such an interval, either supx∈I |Φ(x) − A Kiefer–Wolfowitz theorem 27 Ψ(x)| is attained at an endpoint of I (where Φ = Φ), or it is attained at an interior point, where Ψ < Φ. Since Φ ≤ Φ on [0, 1], it follows that sup |Φ(x) − Ψ(x)| ≤ sup |Φ(x) − Ψ(x)|.

Download PDF sample

Rated 4.53 of 5 – based on 16 votes