Final Exam Calculus I

Date: January 26, 2021 GTIAE Time: 3 hours

Name:
p
1. Compute the area enclosed by the functions f (x) = |1 − x2 | and g(x) = 1 − x in the
interval [0, 1].
√
Solution: Notice that in the interval [0, 1], 1 − x2 > 0 and therefore we can write f (x) = 1 − x2 .
Moreover f (0) = g(0) = 1, f (1) = g(1) = 0 and f (x) > g(x) for x ∈ [0, 1] (you can check so evaluating
at x = 1/2). Therefore the area is given by
Z 1 Z 1 p Z 1p Z 1
(f (x) − g(x)) dx = ( 1 − x2 − (1 − x)) dx = 1 − x2 dx − (1 − x) dx = π/4 − 1/2.
0 0 0 0
where we have used that
 
Z 1p x = sin t, dx = cos t dt Z π/2
1 − x2 dx =  x = 0 → t = arcsin 0 = 0  = cos2 t dt
0 x = 1 → t = arcsin 1 = π/2 0
Z π/2
1 1 sin(2t) t=π/2 π
= 1 + cos(2t) dt = (t + )]t=0 =
2 0 2 2 4
and Z 1
(1 − x) dx = x − x2 /2]x=1
x=0 = 1/2.
0

2. Consider the function Z x2 (t2 +4)

2
F (x) = e−t − e− 2 dt.
0
a) Identify the intervals where the function is increasing/decreasing.
b) Find the maximal interval I ⊂ R that contains the point x = 1 where F is bijective.

Solution: We compute the derivative using Leibniz’s rule:

4 4 +2)/2
F 0 (x) = 2x(e−x − e−(x ).
0
The derivative is continuous on R. We study the sign of
4 4 4
√ the derivative. Using that F (x) = 0 ⇐⇒
x = 0 or − x = −(x + 4)/2 ⇐⇒ x = 4 ⇐⇒ x = ± 2. We have,
 √

 > 0, si x < −√2



 = 0, si x =√− 2
 < 0, si − 2 < x < 0


F 0 (x) = = 0, si x = 0 √
> 0, si 0 < x √ < 2




= 0, si x = −

√ 2




< 0, si x > 2
√ √ √ √
Therefore, F is strictly increasing in (−∞, − 2) ∪ (0,
√ √ 2), strictly decreasing in (− 2, 0) ∪ ( 2, ∞)
and has a minimum at x =√0 and maxima at x = − 2, 2.
b) In particular, I = [0, 2].
Notice that F (x) is an even function, or alternatively, F 0 (x) is an odd function, and therefore, the
above result is obtained just studying the case x > 0.
 √
x3 1 + x6
3. Find the Mclaurin polynomial P17 (x) of degree 17 of f (x) = .
arctan x3
√ x x2 x3 x5
Solution: Using 1 + x = (1 + x)1/2 = 1 + − + O x3 and arctan x = x − + O x7 ,



+
2 8 3 5
 6 12

x3 1 + x2 − x8 + O x18

6 12
1 + x2 − x8 + O x18


f (x) = 9 15 = 6 12 = 1 + a6 x6 + a12 x12 + O(x18 ).
x3 − x3 + x5 + O (x21 ) 1 − x3 + x5 + O (x18 )
Solving recursively
x6 x12 x6 x12
 
+ O x18 = 1 − + O x18 1 + a6 x6 + a12 x12 + O(x18 ) ,




1+ − +
2 8 3 5
5 x6 17 x12
+ O x18 .


one gets f (x) = 1 + −
6 360

4. Prove that 1 + 2 · 2 + 3 · 22 + · · · + n2n−1 = 1 + (n − 1)2n , for n ≥ 1. (Hint: use induction or the

summation formula for a finite geometric series.)

Solution: for n = 1 the summation formula above reads 1 = 1, so it holds. Assume that it holds for
1, . . . , n, and let us check that it also holds for n + 1
1 + 2 · 2 + 3 · 22 + · · · + n2n−1 + (n + 1)2n = 1 + (n − 1)2n + (n + 1)2n = 1 + 2n2n = 1 + n2n+1 ,
which proves that the summation formula holds for all n ≥ 1.
Alternatively, one can deduce the formula from the summation formula for a finite geometric series
1 − xn+1
1 + x + x2 + x3 + · · · + xn = for x 6= 1 and n ≥ 1, by differentiating it
1−x
−(n + 1)xn (1 − x) + 1 − xn+1 1 + (n(x − 1) − 1)xn
1 + 2x + 3x2 + · · · + nxn−1 = = ,
(1 − x)2 (1 − x)2
and evaluating it at x = 2: 1 + 2 · 2 + 3 · 22 + · · · + n2n−1 = 1 + (n − 1)2n .

5. Study the convergence of the improper integral

Z ∞
1 − x sin x
dx
0 1 + ex

Solution: The function f (x) = 1−x sin x

1+ex is continuous on the interval [0, +∞) since the denominator
never vanishes. So, the integral is only improper in +∞. The numerator of the function f (x) changes
sign an infinit number of times when x → +∞. So, we are going to study absolute convergence.
Therefore,
|1 − x sin x| 1+x 1+x
0 ≤ |f (x)| = ≤ ≤ = g(x),
1 + ex 1 + ex ex
for all x ≥ 0. Furthermore,
Z ∞ Z M  
−x u = 1 + x, du = dx
g(x) dx = lim (1 + x)e dx =
0 M →∞ 0 dv = e−x dx, v = −e−x
x=M
= lim −(2 + x)e−x x=0 = lim −(2 + M )e−M + 2 = 2
M →∞ M →∞
R∞
and the integral 0 g(x) dx is convergent. The comparison criteria implies that the improper integral
R +∞ R +∞
0 |f (x)|dx is convergent. So, 0 f (x)dx is absolutely convergent and, therefore, convergent.
6. Study the convergence of the series
∞
X ln n
.
n!
n=1

Solution: we study the convergence of the numerical series by applying the ratio test.
an + 1 ln(n + 1) n! ln(n + 1)
lim = lim = lim =0<1
n→∞ an n→∞ (n + 1)! ln n n→∞ (n + 1) ln(n)

and by the ratio test the series is absolutely convergent and, so, convergent. Notice that to compute
the limit we have used that
ln(n + 1) ln(n + 1) 1
lim = lim · lim =1·0=0
n→∞ (n + 1) ln(n) n→∞ ln n n→∞ n + 1

You can also switch to continuous variable and apply L’Hôpital’s rule.

7. Solve the o.d.e. x + xy 2 dx + y − x2 y dy = 0 and find the solution passing through the point



x = 2, y = 0.

Solution: This o.d.e. P (x, y) dx+Q(x, y) dy = 0 where P (x, y) = x + xy 2 = x 1 + y 2 and Q(x, y) =

∂P ∂Q
y − x2 y = y 1 − x2 is not exact, since



− = 2xy − (−2xy) = 4xy 6≡ 0. To try an integration
∂y ∂x
factor µ(x, y) we need to solve
 
∂P ∂Q ∂µ ∂µ
∂µ
∂µ
− µ=Q −P ⇐⇒ 4xy µ = y 1 − x2 − x 1 + y2 ,
∂y ∂x ∂x ∂y ∂x ∂y
so we search, for instance, for µ depending only on the variable y satisfying the o.d.e. 4yµ =

dµ dµ 4y dy
− 1 + y2 , which is separable = − with solution ln |µ| = −2 ln(1 + y 2 ) + C or
dy µ 1 + y2
K
µ= . Choosing for instance K = 2, we multiply the original o.d.e. by µ to get an equivalent
(1 + y 2 )2
2x 2y(1 − x2 ) x2 − 1
exact o.d.e. dx + dy = 0 which has the general solution = C. The solution
1 + y2 (1 + y 2 )2 1 + y2
x2 − 1
passing through x = 2, y = 0, is = 3.
1 + y2
x y
Alternatively, we can write the o.d.e. as a separable o.d.e. 2
dx + dy = 0, which has
1−x 1 + y2
1 1 1 − x2 −2K x2 − 1 −2K
the solution − ln 1 − x2 + ln 1 + y 2 = K or


= e or = C = ∓e .
2 2 1 + y2 1 + y2

∞
X x2n
8. Find the radius and the interval of convergence of f (x) = (−1)n−1 . What is the
(2n − 1)2n
n=1
value of f (1)?
∞
X x2n−1 x3 x5 x2n+1
Solution: Differentiating, f 0 (x) = (−1)n−1 = x− + − · · · + (−1)n + ··· =
(2n − 1) 3 5 2n + 1
n=1
arctan x, on the interval of convergence I 0 = [−1, 1] and radius of convergence R0 = 1. Therefore f (x)
has also radius of convergence R = 1. For x = ±1, the power series is also convergent by Leibniz
∞ ∞
X 1 X 1
criterion: (−1)n−1 = − (−1)n bn with 0 < bn = → 0 and bn+1 < bn .
(2n − 1)2n (2n − 1)2n
n=1 n=1
Summarizing, the interval of convergence Z of f (x) is I = [−1, 1].
1
Indeed, integrating by parts, f (x) = arctan x dx = x arctan x− ln(1+x2 )+C and 0 = f (0) = C,
2
1 2 π ln 2
so that f (x) = x arctan x − ln(1 + x ), and f (1) = − .
2 4 2