MATH 137 — Fall 2020

Solutions to Written Assignments

and Tips for Writing Good Proofs
Mark Girard

December 8, 2020

The solutions provided by the head instructor provide good explanations for the solutions, but
don’t showcase what a good solution should look like. I’ve compiled here what “good and complete”
solutions to the assignments should look like for you to get an idea of how to write yours.

Contents
Tips for writing good proofs 2

Assignment solutions 3
Assignment 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Assignment 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Assignment 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Assignment 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Assignment 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Assignment 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Assignment 7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Assignment 8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Assignment 9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

1
Tips for writing good proofs
• Clearly state at the beginning what you are trying to prove.

• Start your proof with “Proof :” and end with “QED” or .

• Declare your variables before using them.

– If the statement you are proving has “For all x, P (x)” then in your proof you should
write something like “Let x be given.” Then prove P (x) for the arbitrarily given value
of x.
– If the statement you are proving has “There exists x such that P (x)” then in your proof
you should write something like “Let x be (what ever you choose it to be).” Then Prove
P (x) for this particular value of x.

• Explain your reasoning at each step using words. (You should be able to read your proof
aloud in complete sentences.)

• Clearly state what your assumptions are and when you are using them.

2
 Assignment 1
Let b be a real number. Show that the inequality

|x + 1| + 2|x − 1| < b (1)

has no solutions if and only if b ≤ 2.

Solution
There essentially two things to prove here. The two statements to be proved are:
• If b > 2 then there exists x ∈ R such that |x + 1| + 2|x − 1| < b holds.

• If b ≤ 2 then for every x ∈ R it holds that |x + 1| + 2|x − 1| <

6 b.
We will prove these statements below.
Claim. If b > 2 then the inequality in (1) has a solution.
Proof. Suppose that b > 2. Let x = 1. Then

|x + 1| + 2|x − 1| = |1 + 1| + 2|1 − 1|
=2
<b

and thus the inequality in (1) has a solution.

Claim. If b ≤ 2 then the inequality in (1) does not have a solution.

Proof. Suppose that b ≤ 2 and let x ∈ R be given. There are three cases to consider:
Case 1: Suppose that x < −1. Then |x + 1| = −(x + 1) and |x − 1| = 1 − x, and thus

|x + 1| + 2|x − 1| = −(x + 1) + 2(1 − x)

= 1 − 3x
>4
>2
≥ b,

where we use the fact that x < −1 and thus −3x > 3.

Case 2: Suppose that −1 ≤ x < 1. Then |x + 1| = x + 1 and |x − 1| = 1 − x, and thus

|x + 1| + 2|x − 1| = x + 1 + 2(1 − x)
=3−x
>2
≥ b,

where we use the fact that x < 1 and thus −x > −1.

3
Case 3: Suppose that 1 ≤ x. Then |x + 1| = x + 1 and |x − 1| = x − 1, and thus

|x + 1| + 2|x − 1| = x + 1 + 2(x − 1)
= 3x − 1
≥2
≥ b,

where we use the fact that x ≥ 1 and thus 3x ≥ 3.

In each case, it holds that |x + 1| + 2|x − 1| ≥ b and thus the inequality in (1) has no solutions.

4
Assignment 2
Let a1 , a2 , . . . be a sequence of real numbers such that an ≥ 0 for each n ∈ N, and let L be a
number such that
lim an = L.
n→∞

(a) Prove that L ≥ 0.

2 2
(b) Prove that lim = .
n→∞ an + 5 L+5

Solution.
Part (a)

Claim. It holds that L ≥ 0.

Proof. Towards a contradiction, assume that L < 0 and define  = − L2 (which is positive). As it is
assumed that lim an = L, there exists a number N such that |an − L| <  for every n ≥ N . Hence
n→∞
for every n ≥ N we have that
L L
< an − L < −
2 2
L
which implies that an < 2 and thus an < 0. This contradicts the assumption that an ≥ 0. We
conclude that L ≥ 0.

Part (b)
2 2
Claim. lim = .
n→∞ an + 5 L+5
 
We will prove that: For all  > 0 there exists an N ∈ N such that  an5+2 − 5 
<  for all n ≥ N .

L+2 

Proof. Let  > 0 be given and define 1 = 2(L+2) 5 . Note that 1 > 0 as L ≥ 0. As it is assumed
that the sequence a1 , a2 , . . . converges to L, there exists a number N ∈ N such that |an − L| < 1
for every n ≥ N . Let n ∈ N and suppose that n ≥ N . Note that

|(an + 2)(L + 2)| = (an + 2)(L + 2)

≥ 2(L + 2)
>0

since both an ≥ 0 and L ≥ 0, and thus

1 1
0< ≤ . (2)
|(an + 2)(L + 2)| 2(L + 2)

5
Now
   
 5 5   L − an 
 an + 2 − L + 2  = 5  (an + 2)(L + 2) 
   

|an − L|
=5
|(an + 2)(L + 2)|
5
≤ |an − L| (from (2))
2(L + 2)
5
< 1 (since |an − L| < 1 )
2(L + 2)
2(L+2)
= , (since 1 = 5 )

as desired.

6
 Assignment 3
2
Let a1 , a2 , . . . be the sequence defined by a1 = 1 and an+1 = 4 − an for every n ∈ N. Show that
the sequence converges and compute its limit.

Solution
We will make use of the Monotone Convergence Theorem to prove that the sequence converges. To
do so, we must first show that the sequence is non-decreasing and bounded above. We will prove
the following claim.

Claim. For every n ∈ N it holds that 1 ≤ an ≤ an+1 ≤ 4.

Proof. We proceed by induction.

(Base) Note that a1 = 1 and a2 = 2 and thus the statement is true when n = 1.

(IH) Let k ∈ N and suppose that 1 ≤ ak ≤ ak+1 ≤ 4.

(IS) Note that

1 1 1
1 ≤ ak ≤ ak+1 ≤ 4 =⇒ 1 ≥ ≥ ≥
ak ak+1 4
2 2 1
=⇒ −2≤− ≤− ≤−
ak ak+1 2
2 2 1
=⇒ 2≤4− ≤4− ≤4−
ak ak+1 2
7
=⇒ 2 ≤ ak+1 ≤ ak+2 ≤
2
=⇒ 1 ≤ ak+1 ≤ ak+2 ≤ 4.

From the Principle of Mathematical Induction, we conclude that 1 ≤ an ≤ an+1 ≤ 4 holds for every
n ∈ N.
√
We will now show that the sequence converges to 2 + 2.
√
Claim. lim an = 2 + 2.
n→∞

Proof. From the previous claim, we conclude that the sequence is non-decreasing and bounded
above. It follows from the Monotone Convergence Theorem that the sequence converges. That is,
there is a number L ∈ R such that lim an = L. Moreover the limit L is the least upper bound of
n→∞
the sequence. Therefore a1 ≤ L (that is, 1 ≤ L) and thus L 6= 0. Now

L = lim an = lim an+1

n→∞ n→∞
 
2
= lim 4 −
n→∞ an
2
=4−
L

7
 √ √
and thus L must satisfy
√ L2 − 4L + 2 √
= 0. It follows that either L = 2 + 2 or L = 2 − 2. Observe
that a1 =√1 > 2 − 2 and thus 2 − 2 is not an upper bound of the sequence. We conclude that
L = 2 + 2.

8
Assignment 4
Prove that lim (x3 − 7x + 1) = −5.
x→2

Solution.
We will prove the following statement: For every  > 0 there is a δ > 0 such that for every x ∈ R,
0 < |x − 2| < δ =⇒ |x3 − 7x + 1 − (−5)| < .

Proof. Let  > 0 be given and define δ = min{1, 12 }. Let x ∈ R and suppose that 0 < |x − 2| < δ.
As |x − 2| < 1, it holds that −1 < x − 2 < 1 and thus 1 < x < 3. Hence

0<x−1<2 and 4 < x + 3 < 6,

and thus 0 < |x − 1| < 2 and 0 < |x + 3| < 6. Therefore

0 < |x − 1||x + 3| < 12. (3)


Moreover, it also holds that |x − 2| < 12 . Now

|x3 − 7x + 1 − (−5)| = |x3 − 7x + 6|

= |x − 1||x + 3||x − 2|

< 12
12
= ,

and thus |x3 − 7x + 1 − (−5)| < , as desired.

9
Assignment 5
Suppose c and d are real numbers (with c 6= 0) and consider the function f defined as

cx + 4d
 x<2
2
f (x) = x + 4 2≤x≤3

 2 2x
dx + c + 1 x > 3

for every x ∈ R. For which values of c an d is f continuous everywhere?

Solution.
The function f is continuous everywhere if and only if either c = 2 and d = 1, or c = − 23 and d = 37 .
A plot of the graphs for both cases are shown below.

40
35
30
25
20
15
10
5

1 2 3 4 5

Figure 1: A plot of the graphs of f for different values of c and d. The blue curve is
the graph when c = 2 and d = 1. The red curve is the graph when c = − 23 and d = 73 .

Proof. First note that the function f is clearly continuous on (−∞, 2) ∪ (2, 3) ∪ (3, ∞). It remains
to find the conditions under which f is continuous at x = 2 and x = 3. That is, f is continuous
everywhere if and only if

lim f (x) = f (2) = lim f (x) and lim f (x) = f (3) = lim f (x).
x→2− x→2+ x→3− x→3+

Note that f (2) = 8 and f (3) = 13 and the left- and right-sided limits of f at x = 2 and x = 3 are


lim f (x) = lim cx + 4d = 2c + 4d
x→2 − x→2 −

lim f (x) = lim x2 + 4 = 8 = f (2)

x→2+ x→2+
lim f (x) = lim x2 + 4 = 13 = f (3)


x→3− x→3−
lim f (x) = lim dx2 + 2x 6


and c + 1 = 9d − c + 1.
x→3+ x→3+

10
Therefore, the function f is continuous everywhere if and only if
6
2c + 4d = 8 and 9d − + 1 = 13. (4)
c
Since c 6= 0 we may multiply the second equation by c to see that
6
9d + + 1 = 13 ⇐⇒ 9cd + 6 − 12c = 0
c
⇐⇒ c(3d − 4) + 2 = 0.

The equations in (4) are therefore satisfied if and only if

c = 4 − 2d and c(3d − 4) + 2 = 0. (5)

Plugging c = 4 − 2d into the left-hand side of the second equation, we have

(4 − 2d)(3d − 4) + 2 = −6d2 + 20d − 14 = −2(3d2 − 10 + 7) = −2(3d − 7)(d − 1).

and thus (4 − 2d)(3d − 4) + 2 = 0 if and only if either d = 1 or d = 73 . Plugging these values into
the first equation of (5), we see that f is continuous everywhere if and only if either d = 1 and
c = 2 or d = 37 and c = − 23 .

11
Assignment 6
Suppose a and b are real numbers and consider the function f defined as
(
2x + b x < a
f (x) =
x2 x≥a

for every x ∈ R. For which values of a an b is f differentiable at a?

Solution.
The function f is differentiable at a if and only if a = 1 and b = −1. The graph of the function
with these values is shown below. (Note how the graph smoothly transistions at x = 1.)
10

−1 1 2 3
−2

−4

Proof. Note that f is differentiable at a if and only if the limit

f (a + h) − f (a)
lim
h→0 h
exists. This limit exists if and only if the two one-sided limits
f (a + h) − f (a) f (a + h) − f (a)
lim and lim
h→0− h h→0+ h
exist and their values are the same. Now
f (a + h) − f (a) 2(a + h) + b − a2
lim = lim
h→0− h h→0− h
 (
2a + b − a2 if 2a + b − a2 = 0

2
= lim +2 =
h→0 − h D.N.E. otherwise.

On the other hand,

f (a + h) − f (a) (a + h)2 − a2 2ah + h2

lim = lim = lim = lim 2a + h = 2a.
h→0+ h h→0+ h h→0+ h h→0+

Thus these limits both exist and are the same if and only if 2a + b − a2 = 0 and 2 = 2a. Solving
these equatations, we see that f is differentiable at a if and only if a = 1 and b = −1.

12
Assignment 7
Let f be a differentiable function and let a, b ∈ R such that a 6= b.
(a) Suppose that La (x) = Lb (x) for all x ∈ R. Show that
f (b) − f (a) f 0 (a) + f 0 (b)
= . (6)
b−a 2
(b) Conversely, suppose that (6) holds. Either prove that La (x) = Lb (x) holds for all x ∈ R, or
provide a counterexample if it does not.

Solution.
(a) Recall that the linear approximation functions are defined as
La (x) = f (a) + f 0 (a)(x − a) and Lb (x) = f (b) + f 0 (b)(x − b)
for all x ∈ R. Since La (x) = Lb (x) holds for all x ∈ R, it holds in particular when either
x = a or x = b. Thus, it holds that La (a) = Lb (a) and La (b) = Lb (b). Now,
f 0 (a) + f 0 (b) f 0 (a)(b − a) − f 0 (b)(a − b)
=
2 2(b − a)
La (b) − f (a) − Lb (a) + f (b)
= (since La (b) = f (a) + f 0 (a)(b − a)
2(b − a)
and Lb (a) = f (b) + f 0 (b)(a − b))
Lb (b) − f (a) − La (a) + f (b)
= (since La (a) = Lb (a) and La (b) = Lb (b))
2(b − a)
f (b) − f (a) − f (a) + f (b)
= (since La (a) = f (a) and Lb (b) = f (b))
2(b − a)
2f (b) − 2f (a)
=
2(b − a)
f (b) − f (a)
= ,
b−a
as desired.
(b) Consider the function defined by f (x) = x2 and let a = −1 and b = 1. We have that
f 0 (x) = 2x for all x, and thus
f (b) − f (a) 1−1 f 0 (a) + f 0 (b) (−2) + 2
= =0 and = = 0.
b−a 1 − (−1) 2 2
Hence (6) holds for this function and this choice of a and b. However, the linear approxima-
tions are given by
La (x) = L−1 (x) = 1 − 2(x + 1) and Lb (x) = L1 (x) = 1 + 2(x − 1)
for all x ∈ R. These are not the same function, since in particular we have that
La (1) = 1 − 2(1 + 1) = −3 but Lb (1) = 1 + 2(1 − 1) = 0
and thus La (1) 6= Lb (1).

13
Assignment 8
√ √ √
Let c > 0 be given and consider the curve defined by the equation x + y = c. Let L be a
tangent line to the curve with exactly one x-intercept and exactly one y-intercept. Prove that the
sum of the x- and y-intercepts of L is equal to c.

Solution.
√ √ √
Proof. Let (x0 , y0 ) be the point on the curve (so that x0 + y0 = c) at which the line L is
tangent, and let m be the slope of this line such that

L = {(x, y) ∈ R2 : y = y0 + m(x − x0 )}.

(Note that we cannot have either x0 = 0 or y0 = 0, since otherwise the tangent line would have
either infinitely many y-intercepts or infinitely many x-intercepts.)
We may use the method of implicit differentiation to find the slope of this line. On the curve,
we have that
d√ d √ √
1 1 dy
0= c= x+ y = √ + √
dx dx 2 x 2 y dx
and thus √
dy y
= −√ .
dx x
The slope of the tangent line at (x0 , y0 ) may be expressed as
 √  √
dy  y  y0
m= = −√  = −√ .
dx (x,y)=(x0 ,y0 )
 x (x,y)=(x0 ,y0 ) x0

Let x1 and y1 denote the x- and y-intercepts of the line L, respectively. These are the values such
that (x1 , 0) and (0, y1 ) are on the line. That is, the values x1 and y1 must satisfy

y1 = y0 − mx0 and 0 = y0 + m(x1 − x0 ).

Rearranging, we find that

√
y0 √
y1 = y0 − mx0 = y0 + √ x0 = y0 + x0 y0
x0

and √
1 x0 √
x1 = x0 − y0 = x0 + √ y0 = x0 + x0 y0 .
m y0
Summing the x- and y-intercepts, we find that
√ √ √ √ 2
x1 + y1 = x0 + y0 + 2 x0 y0 = ( x0 + y0 )2 = c = c,

as desired.

14
Assignment 9
Let L ∈ R with L 6= 0. Suppose that f is a differentiable function such that lim f (x) = L and
x→∞
that lim f 0 (x) exists. Prove that lim f 0 (x) = 0 using two different methods:
x→∞ x→∞

(a) using l’Hopital’s rule.

(b) using the Mean Value Theorem.

Solution.
(a) If L < 0 then lim ex f (x) = −∞. On the other hand, if L > 0 then lim ex f (x) = +∞. In
x→∞ x→∞
x x ex f (x)
either case, we have that lim e f (x) = ±∞ and that lim e = +∞, and thus lim
x→∞ x→∞ x→∞ ex
is an indefinite form of type ± ∞
∞ . We may therefore apply l’Hopital’s rule to find that

ex f (x)
L = lim f (x) = lim
x→∞ x→∞ ex
e (f (x) + f 0 (x))
x
= lim by l’Hopital’s rule
x→∞ ex
= lim (f (x) + f 0 (x))
x→∞
= lim f (x) + lim f 0 (x)
x→∞ x→∞
0
= L + lim f (x),
x→∞

and thus lim f 0 (x) = 0.

x→∞

(b) We may construct an infinite sequence {cn }n∈N as follows. For every n ∈ N, the Mean Value
Theorem allows us to find a point cn ∈ (n, n + 1) satisfying

f (n + 1) − f (n)
f 0 (cn ) = = f (n + 1) − f (n).
n+1−n
Note that n < cn for every n ∈ N and thus limn→∞ cn = ∞. By the Sequential Characteriza-
tion of Limits and the fact that lim f 0 (x) exists, we have that
x→∞

lim f 0 (x) = lim f 0 (cn )

x→∞ n→∞
= lim (f (n + 1) − f (n))
n→∞
= lim f (n + 1) − lim f (n)
n→∞ n→∞
=L−L
= 0,

as desired.

15