A functional limit theorem for the profile of search trees

From MaRDI portal
Publication:2476407



Abstract: We study the profile Xn,k of random search trees including binary search trees and m-ary search trees. Our main result is a functional limit theorem of the normalized profile Xn,k/mathbbEXn,k for k=lflooralphalognfloor in a certain range of alpha. A central feature of the proof is the use of the contraction method to prove convergence in distribution of certain random analytic functions in a complex domain. This is based on a general theorem concerning the contraction method for random variables in an infinite-dimensional Hilbert space. As part of the proof, we show that the Zolotarev metric is complete for a Hilbert space.


The level profile of a tree, as examined in this paper, is the sequence of numbers each counting the number of nodes with the same distance from the root. This parameter is a very informative shape parameter, useful for many purposes. The authors derive a functional limit law of the form \[ \frac{X_{n,\lfloor \alpha\log n\rfloor}} {E(X_{n,\lfloor \alpha\log n\rfloor})} \to X(\alpha), \] for the profile \(X_{n,k}\) of a class of trees called the generalized \(m\)-ary search trees. Many deep tools are developed for achieving this, which are of independent interest \textit{per se}.



Cites work


Cited in
(32)








This page was built for publication: A functional limit theorem for the profile of search trees

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2476407)