This document provides solutions to 5 problems on a MAT319/320 midterm exam.
1) It defines the limit of a sequence as requiring any epsilon neighborhood of the limit L to eventually contain all terms of the sequence.
2) It proves that the union of two bounded sets is bounded, and that the supremum of the union is the maximum of the individual supremums.
3) It uses induction to prove 2^n is greater than or equal to n+1 for all natural numbers n, and also proves 2^n is greater than or equal to n^2 for n greater than or equal to 4.
4) It proves the limit of the sequence a_n = 1/2
This document provides solutions to 5 problems on a MAT319/320 midterm exam.
1) It defines the limit of a sequence as requiring any epsilon neighborhood of the limit L to eventually contain all terms of the sequence.
2) It proves that the union of two bounded sets is bounded, and that the supremum of the union is the maximum of the individual supremums.
3) It uses induction to prove 2^n is greater than or equal to n+1 for all natural numbers n, and also proves 2^n is greater than or equal to n^2 for n greater than or equal to 4.
4) It proves the limit of the sequence a_n = 1/2
This document provides solutions to 5 problems on a MAT319/320 midterm exam.
1) It defines the limit of a sequence as requiring any epsilon neighborhood of the limit L to eventually contain all terms of the sequence.
2) It proves that the union of two bounded sets is bounded, and that the supremum of the union is the maximum of the individual supremums.
3) It uses induction to prove 2^n is greater than or equal to n+1 for all natural numbers n, and also proves 2^n is greater than or equal to n^2 for n greater than or equal to 4.
4) It proves the limit of the sequence a_n = 1/2
This document provides solutions to 5 problems on a MAT319/320 midterm exam.
1) It defines the limit of a sequence as requiring any epsilon neighborhood of the limit L to eventually contain all terms of the sequence.
2) It proves that the union of two bounded sets is bounded, and that the supremum of the union is the maximum of the individual supremums.
3) It uses induction to prove 2^n is greater than or equal to n+1 for all natural numbers n, and also proves 2^n is greater than or equal to n^2 for n greater than or equal to 4.
4) It proves the limit of the sequence a_n = 1/2
1. 10 points Give a careful and complete definition of what it means when we say “The limit of the sequence X is L.”
Solution: The limit of the sequence X = (xn ) is L if for any > 0, there is a natural number K so that |xn − L| < for all n ≥ K .
2. 10 points Let A and B be bounded subsets of R.
(a) Prove that A ∪ B is a bounded subset of R. Solution: Since A is bounded, there are real numbers uA and lA so that every element of A lies between uA and lA ; that is, A ⊆ [lA , uA ]. Similarly, there are real numbers lB and uB so that B ⊆ [lB , uB ]. Let L = inf(lA , lB ), and let U = sup(uA , uB ). Then certainly A ⊆ [L, U ], and also B ⊆ [L, U ]. If x ∈ A ∪ B, then x ∈ A or x ∈ B, and so x ∈ [L, U ]. Thus A ∪ B ⊆ [L, U ], which means A ∪ B is bounded.
(b) Prove that sup(A ∪ B) = sup(sup A, sup B).
Solution: Let U = sup(sup A, sup B). We must show that for any x ∈ A ∪ B, we have x ≤ U , and we must also show that if V < U , then there is a y ∈ A ∪ B so that y > V . The first part is nearly immediate: since U ≥ sup A, for every a ∈ A we have a ≤ U . Similarly, since U ≥ sup B, we have b ≤ U for every b ∈ B. For the second part, note that either U = sup A or U = sup B. Without loss of generality, suppose the former holds. Then since U is the supremum of A, if V < U , there is an a ∈ A so that a > V . Since this same a is an element of A ∪ B, we have the desired conclusion.
3. (a) 10 points Prove that for all natural numbers n, we have 2n ≥ n + 1.
You might find induction helpful. Solution: First, we see that for n = 1, we have 2 = 21 ≥ 1 + 1, so the base case holds. Now for the inductive step, we want to show that 2k ≥ k + 1 implies that 2k+1 ≥ k + 2. We have
= 2k + 2 > k + 2, Page 2 of 3 where we have used k > 0 in the final step. This shows that 2n ≥ n + 1 for all natural numbers n.
(b) Prove that for all natural numbers n ≥ 4, we have 2n ≥ n2 .
Feel free to use the result from part a, even if you couldn’t do it. Solution: First, we establish the relation for n = 4, our base case. We have 24 = 16 = 42 , as desired. Now we show that if 2k ≥ k 2 , then we also have 2k+1 ≥ (k + 1)2 . Unfortunately, it seems we need to establish that 2k ≥ 2k + 1 first, so let’s do that, again by induction. First, we see that 24 = 16 > 9 = 2 · 4 + 1. For the inductive step, we must establish that 2k+1 ≥ 2(k + 1) + 1. So
2k+1 = 2 · 2k ≥2 · (2k + 1) using the induction hypothesis
=4k + 2 = 2k + 2k + 2.
Since k > 1, we have 2k > 2 and so we have
2k + 2k + 2 > 2k + 4 > 2(k + 1) + 1.
This is our desired result.
Now we return to our main proof:
2k+1 = 2k + 2k ≥ k 2 + (2k + 1) = (k + 1)2 .
Here we used the induction hypothesis (2k ≥ k 2 ) and our previous fact (2k ≥ 2k + 1). I’m sorry that a typo in part a caused it not to be useful (although still true).
n −1 4. 10 points Consider the sequence whose nth term an = . Prove, using the 2 definition of the limit, that the limit of this sequence is 0.
Solution: Given > 0, we need to find a K so that |an | < for all n ≥ K. Note that from the previous problem, we know that 2n ≥ n + 1 > n, and so if we take K ≥ 1/, for any n > K we will have n −1 = 1 ≤ 1 ≤ 1 = , 2 2n n 1/
as desired.
MAT319/320: Solutions to First Midterm October 8, 2008
Page 3 of 3 5. 10 points Let f : (0, 1) → R have the property that f (x) < x for all x ∈ (0, 1). (a) Prove that sup f (x) ≤ 1 x∈(0,1)
Solution: Suppose not. If the supremum is greater than 1, there must be some some x ∈ (0, 1) with 1 < f (x). But f (x) < x < 1, and so this is a contradiction.
(b) Is it true that sup f (x) < 1? Prove or give a counterexample.
x∈(0,1)
Solution: This is not true.
There are plenty of counterexamples, such as f (x) = x2 or f (x) = 2x − 1. Let’s use the second one. For f (x) = 2x − 1, we certainly have f (x) < x for all x < 1. But the image of the interval (0, 1) under f is (−1, 1), and so sup f (x) = 1.
MAT319/320: Solutions to First Midterm October 8, 2008
