This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Statistics 611: Homework 2 (1) (i) Let G n be a sequence of random functions defined on R p with continuous sample paths: i.e. for each ω , G ( ω ) n ( t ) is continuous in t . Suppose that g is a fixed function on R p . All functions are taken to be realvalued. Assume that (i) g is maximized uniquely at t ? , (ii) For every compact subset K of R p , k G n g k ∞ ,K ≡ sup t ∈ K  G n ( x ) g ( x ) → p 0. (iii) Let { T n } be a sequence of random variables maximizing G n such that T n is tight i.e. bounded in probability. Show that T n converges in probability to t ? . (ii) Consider now, a sequence of random functions G n on C ( K ) where K is a compact subset of R , converging to g ∈ C ( K ). Suppose that { T n } solves G n ( T n ) = 0 and that g ( t ) = 0 for a unique t ∈ K . Show that T n → p t . (2) The GlivenkoCantelli Theorem. Let X 1 ,X 2 ,..., be an i.i.d. sequence of random variables with common distribution F on R . Let F n denote the empirical c.d.f. of the first n observations X 1 ,X 2 ,...,X n . Thus, F n ( x ) = n 1 ∑ n i =1 1( X i ≤ x ). We seek to show that sup x  F n ( x ) F ( x ) → a.s 0. To this end, fix an arbitrary > 0, and choose points∞ = x < x 1 ≤ x 2 ≤ x 3 ≤ ... ≤ x k 1 < x k = ∞ , such that (a) k >...
View
Full
Document
This note was uploaded on 04/14/2010 for the course STATS 612 taught by Professor Moulib during the Winter '08 term at University of Michigan.
 Winter '08
 moulib
 Statistics

Click to edit the document details