{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

psAnalysis3Ans

# psAnalysis3Ans - Fall 2007 ARE211 Problem Set#03 Answer key...

This preview shows pages 1–3. Sign up to view the full content.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Fall 2007 ARE211 Problem Set #03- Answer key Third Analysis Problem Set Problem 1 For the following problem, consider an arbitrary universe X and an arbitrary metric d defined on X × X . State whether the following statements are true or false. If they are true, give a proof. If they are wrong, give a counter-example a) If a sequence x n converges then every subsequence of x n must also converge. b) If every subsequence of the sequence x n converges, then x n must also converge. c) If every subsequence of the sequence x n converges, then they all converge to the same point. d) If one subsequence of the sequence x n converges, then x n must also converge. Ans: Recall the definitions (C) Convergence : A sequence x n in X converges to an element ¯ x ∈ X in the metric d if ∀ epsilon1 > ∃ N ∈ N such that ∀ n > N we have d ( x n , ¯ x ) < epsilon1 (S) Subsequence : A sequence { y 1 ,y 2 ,...,y n ,... } is a subsequence of another sequence if there exists a strictly increasing function τ : N → N such that for all n ∈ N : y n = x τ ( n ) Let’s first prove the following lemma (L) by induction: ∀ strictly increasing τ : N → N we must have τ ( n ) ≥ n ∀ n ∈ N (i) initialization: for n = 1: τ ( n ) ≥ 1 as τ maps from N into N . (ii) induction hypothesis: ∀ n ≤ ¯ n : τ ( n ) ≥ n . (iii) induction step: ¯ n → ¯ n + 1 τ (¯ n + 1) > τ (¯ n ) (as τ () is strictly increasing) ≥ ¯ n (From (ii)). Since τ (¯ n + 1) ∈ N we therefore know that τ (¯ n + 1) ≥ ¯ n + 1 . Let’s consider the problems: a) True: If a sequence x n converges then so does every subsequence of x n . (1) You are given that a sequence x n converges, i.e., ∃ ¯ x ∈ X s.t. ∀ epsilon1 > ∃ N ∈ N such that ∀ n > N we have d ( x n , ¯ x ) < epsilon1 (2) Using lemma (L) in the definition of a subsequence (S) we know that ∀ n : y n = x τ ( n ) = x ˆ n for some ˆ n ≥ n . (3) But (2) and (1) imply that ∀ epsilon1 > ∃ N ∈ N such that ∀ n > N we have d ( y n , ¯ x ) < epsilon1 . Hence y n converges to ¯ x . b) True: If every subsequence of the sequence x n converges then so does x n . This is trivial: Since every subsequence convergence, pick the subsequence τ ( n ) = n which is simply the sequence itself. Hence the sequence converges. c) True: from b), ( x n ) itself converges. Let ¯ x be the point to which it converges. Fix epsilon1 > ; ∃ N epsilon1 ∈ N such that ∀ n > N epsilon1 , d ( x n ,x ) < epsilon1 . Now consider an arbitrary subsequence ( y n ) of ( x n ) and let τ ( · ) denote the strictly increasing function from N to N that defines ( y n ) . From class we know that τ ( · ) increases without bound. Hence there exists M ∈ N , such that for all m > M , τ ( m ) > N epsilon1 so that y m = x τ ( m ) ∈ B (¯ x,epsilon1 ) . Hence ( y n ) converges to ¯ x . 2 Problem 2 For the following problem, consider an arbitrary universe X and an arbitrary metric d defined on...
View Full Document

{[ snackBarMessage ]}

### Page1 / 7

psAnalysis3Ans - Fall 2007 ARE211 Problem Set#03 Answer key...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online