0% found this document useful (0 votes)
30 views

Test 1 Sol

This document contains solutions to test questions for the course MATH2822 Mathematical Methods for Actuarial Science II. The solutions involve finding planes containing lines, determining values of constants in partial derivative equations, applying the second derivative test to find properties of functions at local extrema, using linear approximations, evaluating limits involving multivariable functions, and solving an implicit differentiation problem.

Uploaded by

ching chau
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
30 views

Test 1 Sol

This document contains solutions to test questions for the course MATH2822 Mathematical Methods for Actuarial Science II. The solutions involve finding planes containing lines, determining values of constants in partial derivative equations, applying the second derivative test to find properties of functions at local extrema, using linear approximations, evaluating limits involving multivariable functions, and solving an implicit differentiation problem.

Uploaded by

ching chau
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 11

THE UNIVERSITY OF HONG KONG

DEPARTMENT OF MATHEMATICS

MATH2822 Mathematical Methods for Actuarial Science II


Test 1 Solutions

1. (Version A) (2 points) Which of the following planes contains the straight line (3, 1, −2) +
(7, −2, −4)t?

(a) 2x + 3y + 2z = 5
(b) −7x + 2y + 4z = −27
(c) 3x + y − 2z = 27
(d) x + 3y + 3z = −11

Solution. Answer: (a)

Since the line contains the point (3, 1, −2) and is parallel to the vector (7, −2, −4), the point
(3, 1, −2) lies on the plane, and the vector (7, −2, −4) is perpendicular to the normal of the
plane. We check that 2(3) + 3(1) + 2(−2) = 5 and (2, 3, 2) ⋅ (7, −2, −4) = 0. So 2x + 3y + 2z = 5 is
a plane containing the given straight line. The other planes do not have the same properties.

(Version B) (2 points) Which of the following planes contains the straight line (4, −3, 5) +
(2, −5, 1)t?

(a) 2x + y + z = 10
(b) −2x + 5y − z = −28
(c) 4x − 3y + 5z = 28
(d) 3x − y − 3z = 8

Solution. Answer: (a)

Since the line contains the point (4, −3, 5) and is parallel to the vector (2, −5, 1), the point
(4, −3, 5) lies on the plane, and the vector (2, −5, 1) is perpendicular to the normal of the
plane. We check that 2(4) + (−3) + (5) = 10 and (2, 1, 1) ⋅ (2, −5, 1) = 0. So 2x + y + z = 10 is a
plane containing the given straight line. The other planes do not have the same properties.

1
2. (Version A) (2 points) Let c be a constant, and let f (x, y) = sin (x + cy). Given that fxy (x, y) =
4f (x, y), find the value of c.

(a) 4
(b) −4
(c) 2
(d) −2

Solution. Answer: (b)

We have fx (x, y) = cos (x + cy) and fxy (x, y) = −c sin (x + cy). It is given that −c sin (x + cy) =
4 sin (x + cy), so c = −4.

(Version B) (2 points) Let c be a constant, and let f (x, y) = cos (x + cy). Given that fxy (x, y) =
6f (x, y), find the value of c.

(a) 6
(b) −6
(c) 3
(d) −3

Solution. Answer: (b)

We have fx (x, y) = − sin (x + cy) and fxy (x, y) = −c cos (x + cy). It is given that −c cos (x + cy) =
6 cos (x + cy), so c = −6.

2
3. (Version A) (2 points) Let f (x, y) be a scalar-valued function whose first-order and second-
order partial derivatives are continuous. Suppose (a, b) is a point of local minimum of f (x, y)
such that fxy (a, b) ≠ 0. Which of the following must be correct?

(a) fxx (a, b) > 0 and fyy (a, b) > 0


(b) fxx (a, b) > 0 > fyy (a, b)
(c) fxx (a, b)fyy (a, b) ⩽ fxy (a, b)2
(d) fxx (a, b)fyy (a, b) ≠ fxy (a, b)2

Solution. Answer: (a)


2
(a) ✓: By the second partial derivatice test, we must have fxx fyy −fxy ⩾ 0, or otherwise (a, b)
2
is a saddle point. This implies fxx fyy ⩾ fxy > 0. So either fxx , fyy > 0 or fxx , fyy < 0.
The latter case is rejected by the second partial derivative test, since otherwise (a, b) is a
point of local maximum.
(b) ×: See above.
(c) ×: E.g. f (x, y) = x2 + xy + y 2 has a local minimum at (0, 0), but fxx fyy > fxy
2
.
(d) ×: E.g. f (x, y) = (x + y)2 has a local minimum at (0, 0), but fxx fyy = fxy
2
.

(Version B) (2 points) Let f (x, y) be a scalar-valued function whose first-order and second-
order partial derivatives are continuous. Suppose (a, b) is a point of local maximum of f (x, y)
such that fxy (a, b) ≠ 0. Which of the following must be correct?

(a) fxx (a, b) < 0 and fyy (a, b) < 0


(b) fxx (a, b) < 0 < fyy (a, b)
(c) fxx (a, b)fyy (a, b) ⩽ fxy (a, b)2
(d) fxx (a, b)fyy (a, b) ≠ fxy (a, b)2

Solution. Answer: (a)


2
(a) ✓: By the second partial derivatice test, we must have fxx fyy −fxy ⩾ 0, or otherwise (a, b)
2
is a saddle point. This implies fxx fyy ⩾ fxy > 0. So either fxx , fyy > 0 or fxx , fyy < 0.
The former case is rejected by the second partial derivative test, since otherwise (a, b) is
a point of local minimum.
(b) ×: See above.
(c) ×: E.g. f (x, y) = −(x2 + xy + y 2 ) has a local maximum at (0, 0), but fxx fyy > fxy
2
.
(d) ×: E.g. f (x, y) = −(x + y)2 has a local maximum at (0, 0), but fxx fyy = fxy
2
.

3

4. (Version A) (3 points) Let f (x, y) = x(x + y). Find the approximation to f (1.4, 3.5) using
the linearization of f at the point (1, 3). (Correct the answer to 2 decimal places.)

Solution. Answer: 2.63


2x + y x
We have fx (x, y) = √ and fy (x, y) = √ . So the linearization of f at (1, 3)
2 x(x + y) 2 x(x + y)
is
5 1
f (1, 3) + fx (1, 3)(x − 1) + fy (1, 3)(y − 3) = 2 + (x − 1) + (y − 3).
4 4
The approximation to f (1.4, 3.5) is
5 1
2 + (1.4 − 1) + (3.5 − 3) = 2.63.
4 4


(Version B) (3 points) Let f (x, y) = y(x + y). Find the approximation to f (6.7, 2.6) using
the linearization of f at the point (6, 2). (Correct the answer to 2 decimal places.)

Solution. Answer: 4.93


y x + 2y
We have fx (x, y) = √ and fy (x, y) = √ . So the linearization of f at (6, 2)
2 y(x + y) 2 y(x + y)
is
1 5
f (6, 2) + fx (6, 2)(x − 6) + fy (6, 2)(y − 2) = 4 + (x − 6) + (y − 2).
4 4
The approximation to f (6.7, 2.6) is
1 5
4 + (6.7 − 6) + (2.6 − 2) = 4.93.
4 4

4
xa y a+3
5. (Version A) (3 points) Find the smallest positive integer a such that lim = 0.
(x,y)→(0,0) (x2 + y 2 )5

Solution. Answer: 4

Using polar coordinates, we find that

xa y a+3 (r cos θ)a (r sin θ)a+3


lim = lim = lim r2a−7 cosa θ sina+3 θ.
(x,y)→(0,0) (x2 + y 2 )5 r→0 r10 r→0

Since cosa θ sina+3 θ is bounded, the limit is 0 when a = 4. For a ⩽ 3, the exponent of r is
negative, so the limit does not exist.

xa+3 y a
(Version B) (3 points) Find the smallest positive integer a such that lim = 0.
(x,y)→(0,0) (x2 + y 2 )7

Solution. Answer: 6

Using polar coordinates, we find that

xa+3 y a (r cos θ)a+3 (r sin θ)a


lim 2 2 7
= lim = lim r2a−11 cosa+3 θ sina θ.
(x,y)→(0,0) (x + y ) r→0 r14 r→0

Since cosa+3 θ sina θ is bounded, the limit is 0 when a = 6. For a ⩽ 5, the exponent of r is
negative, so the limit does not exist.

5
6. (Version A) (3 points) Suppose w and x are independent variables, y = f (x) is a differentiable
function in x, and z = g(x, y) is a differentiable function in x and y such that x2 z + yez = w.
Given that f (2) = 1, g(2, 1) = 0, gx (2, 1) = 3 and gy (2, 1) = 5, compute f ′ (2).

15
Solution. Answer: −
26

dz dx dy
By the chain rule, we have = gx (x, y) + gy (x, y) = gx (x, y) + f ′ (x)gy (x, y). Putting
dx dx dx
dz
x = 2 and y = 1, we obtain = gx (2, 1) + f ′ (2)gy (2, 1) = 3 + 5f ′ (2) at (x, y, z) = (2, 1, 0). Now,
dx
we differentiate both sides of x2 z + yez = w with respect to x. This gives
dz dy z dz
2xz + x2 + e + yez = 0.
dx dx dx
Putting (x, y, z) = (2, 1, 0), this yields

4(3 + 5f ′ (2)) + f ′ (2) + (3 + 5f ′ (2)) = 0.


15
Solving this, we get f ′ (2) = − .
26

(Version B) (3 points) Suppose w and x are independent variables, y = f (x) is a differentiable


function in x, and z = g(x, y) is a differentiable function in x and y such that x2 z + yez = 3w.
Given that f (2) = 3, g(2, 3) = 0, gx (2, 3) = 4 and gy (2, 3) = 3, compute f ′ (2).

14
Solution. Answer: −
11

dz dx dy
By the chain rule, we have = gx (x, y) + gy (x, y) = gx (x, y) + f ′ (x)gy (x, y). Putting
dx dx dx
dz
x = 2 and y = 3, we obtain = gx (2, 3) + f ′ (2)gy (2, 3) = 4 + 3f ′ (2) at (x, y, z) = (2, 3, 0). Now,
dx
we differentiate both sides of x2 z + yez = 3w with respect to x. This gives
dz dy z dz
2xz + x2 + e + yez = 0.
dx dx dx
Putting (x, y, z) = (2, 3, 0), this yields

4(4 + 3f ′ (2)) + f ′ (2) + 3(4 + 3f ′ (2)) = 0.


14
Solving this, we get f ′ (2) = − .
11

6
7. (Version A) Let k be a constant. Let ℓ1 be the straight line (−2, 4, −3)+(1, 2, 3)t, and let ℓ2 be
the straight line (k, −2, 3) + (−2, −5, k)t. Suppose ℓ1 and ℓ2 are perpendicular to each other.

(a) (3 points) Find the value of k.


(b) (4 points) Find the equation of the plane passing through the point (3, 1, 0) and is parallel
to both ℓ1 and ℓ2 . (Express your answer in the form Ax + By + Cz = D.)

Solution.
(a) Answer: 4

Since ℓ1 and ℓ2 are perpendicular, we have (1, 2, 3) ⋅ (−2, −5, k) = 0. So −12 + 3k = 0, and
hence k = 4.

(b) As the lines are parallel to the vectors (1, 2, 3) and (−2, −5, 4) respectively, a normal to
the plane is (1, 2, 3) × (−2, −5, 4) = (23, −10, −1). By proposition 1.10, the equation of the
plane is
23(x − 3) − 10(y − 1) − (z − 0) = 0,
i.e. 23x − 10y − z = 59.

(Version B) Let k be a constant. Let ℓ1 be the straight line (3, 1, 4) + (2, 4, 1)t, and let ℓ2 be
the straight line (k, −1, 4) + (−3, 2, k)t. Suppose ℓ1 and ℓ2 are perpendicular to each other.

(a) (3 points) Find the value of k.


(b) (4 points) Find the equation of the plane passing through the point (2, 3, 0) and is parallel
to both ℓ1 and ℓ2 . (Express your answer in the form Ax + By + Cz = D.)

Solution.

(a) Answer: −2

Since ℓ1 and ℓ2 are perpendicular, we have (2, 4, 1) ⋅ (−3, 2, k) = 0. So 2 + k = 0, and hence


k = −2.

(b) As the lines are parallel to the vectors (2, 4, 1) and (−3, 2, −2) respectively, a normal to
the plane is (2, 4, 1) × (−3, 2, −2) = (−10, 1, 16). By proposition 1.10, the equation of the
plane is
−10(x − 2) + (y − 3) + 16(z − 0) = 0,
i.e. −10x + y + 16z = −17.

7
8. (Version A) Let S be the graph of a function f (x, y). The equation of S in spherical coordi-
π
nates is ρ2 = 2ρ sin φ sin θ + 3 with 0 ⩽ φ ⩽ .
2
(a) (4 points) Find the equation of S in Cartesian coordinates and describe this surface.
(b) (1 point) Write down a formula for f (x, y). (No explanation is needed.)

(c) (4 points) Find the equation of the tangent plane to S at (x, y, z) = (1, 2, 2). (Express
your answer in the form Ax + By + Cz = D.)

Solution.
(a) By proposition 1.13, we have

ρ2 = 2ρ sin φ sin θ + 3
⇔ x2 + y 2 + z 2 = 2y + 3
⇔ x2 + (y − 1)2 + z 2 = 4.
π
The condition 0 ⩽ φ ⩽ means the points lie above the xy-plane. Therefore, S is the set
2
of all points satisfying x2 + (y − 1)2 + z 2 = 4 and z ⩾ 0. It is the upper hemisphere with
centre (0, 1, 0) and radius 2.

(b) f (x, y) = 4 − x2 − (y − 1)2

x y−1
(c) We have fx (x, y) = − √ and fy (x, y) = − √ . By definition
4 − x2
− (y − 1)2 4 − x2 − (y − 1)2
2.16, the equation of the tangent plane is

z = f (1, 2) + fx (1, 2)(x − 1) + fy (1, 2)(y − 2).


√ 1 1 √
This means z = 2 − √ (x − 1) − √ (y − 2), or x + y + 2z = 5.
2 2

8
(Version B) Let S be the graph of a function f (x, y). The equation of S in spherical coordinates
π
is ρ2 = 2ρ sin φ cos θ + 8 with 0 ⩽ φ ⩽ .
2
(a) (4 points) Find the equation of S in Cartesian coordinates and describe this surface.
(b) (1 point) Write down a formula for f (x, y). (No explanation is needed.)

(c) (4 points) Find the equation of the tangent plane to S at (x, y, z) = (2, 1, 7). (Express
your answer in the form Ax + By + Cz = D.)

Solution.
(a) By proposition 1.13, we have

ρ2 = 2ρ sin φ cos θ + 8
⇔ x2 + y 2 + z 2 = 2x + 8
⇔ (x − 1)2 + y 2 + z 2 = 9.
π
The condition 0 ⩽ φ ⩽ means the points lie above the xy-plane. Therefore, S is the set
2
of all points satisfying (x − 1)2 + y 2 + z 2 = 9 and z ⩾ 0. It is the upper hemisphere with
centre (1, 0, 0) and radius 3.

(b) f (x, y) = 9 − (x − 1)2 − y 2

x−1 y
(c) We have fx (x, y) = − √ and fy (x, y) = − √ . By definition
9 − (x − 1)2 − y2 9 − (x − 1)2 − y 2
2.16, the equation of the tangent plane is

z = f (2, 1) + fx (2, 1)(x − 2) + fy (2, 1)(y − 1).


√ 1 1 √
This means z = 7 − √ (x − 2) − √ (y − 1), or x + y + 7z = 10.
7 7

9
3x2 + 4xy
9. (Version A) Define f (x, y) = for 0 < x2 + y 2 ⩽ 4. It can be shown that fx (x, y) =
x2 + y 2
−2y(x − 2y)(2x + y) 2x(x − 2y)(2x + y)
and fy (x, y) = .
(x2 + y 2 )2 (x2 + y 2 )2

(a) (3 points) Is it possible to define f (0, 0) so that f is continuous at (0, 0)?


(b) (6 points) Find the global extrema of f . (You should show the use of calculus in solving
this question. You may assume the global extrema exist.)

Solution.
3x2 + 4xy
(a) No. It suffices to check lim does not exist. Considering the restriction
(x,y)→(0,0) x2 + y 2
0
of the function to x = 0, the limit becomes lim 2 = 0. Considering the restriction of
y→0 y
3x2
the function to y = 0, the limit becomes lim 2 = 3. Since they do not agree with each
x→0 x
other, the limit does not exist.

(b) For interior points, we solve ∇f (x, y) = (0, 0). This means

−2y(x − 2y)(2x + y) = 2x(x − 2y)(2x + y) = 0.

If x = 0 or y = 0, it is clear that x = y = 0. This is rejected since (0, 0) is not in the


domain of f . If x, y ≠ 0, both equations are the same as (x − 2y)(2x + y) = 0. This gives
x = 2y or y = −2x. So all points (2t, t), (t, −2t) are critical points.

For boundary points with g(x, y) = x2 + y 2 = 4, we use the method of Lagrange multiplier
to optimize h(x, y) = 4f (x, y) = 3x2 + 4xy. Note that ∇g(x, y) = (2x, 2y) ≠ (0, 0) since
x = y = 0 is the only solution but 02 + 02 ≠ 4. Thus, we solve ∇h(x, y) = λ∇g(x, y), i.e.

6x + 4y = 2λx,
{
4x = 2λy.

Combining the equations, we obtain 6xy + 4y 2 = 2λxy = 4x2 . This is again the same as
2(x − 2y)(2x + y) = 0, and so the solutions are (2t, t), (t, −2t) (one can solve for t using
x2 + y 2 = 4, but this is not needed).

As f (2t, t) = 4 and f (t, −2t) = −1, we see that the global maximum of f is 4 and the
global minimum is −1.

10
3xy − 4y 2
(Version B) Define f (x, y) = for 0 < x2 + y 2 ⩽ 9. It can be shown that fx (x, y) =
x2 + y 2
−y(x − 3y)(3x + y) x(x − 3y)(3x + y)
and fy (x, y) = .
(x2 + y 2 )2 (x2 + y 2 )2

(a) (3 points) Is it possible to define f (0, 0) so that f is continuous at (0, 0)?


(b) (6 points) Find the global extrema of f . (You should show the use of calculus in solving
this question. You may assume the global extrema exist.)

Solution.
3xy − 4y 2
(a) No. It suffices to check lim does not exist. Considering the restriction
(x,y)→(0,0) x2 + y 2
−4y 2
of the function to x = 0, the limit becomes lim 2 = −4. Considering the restriction of
y→0 y
0
the function to y = 0, the limit becomes lim 2 = 0. Since they do not agree with each
x→0 x
other, the limit does not exist.

(b) For interior points, we solve ∇f (x, y) = (0, 0). This means

−y(x − 3y)(3x + y) = x(x − 3y)(3x + y) = 0.

If x = 0 or y = 0, it is clear that x = y = 0. This is rejected since (0, 0) is not in the


domain of f . If x, y ≠ 0, both equations are the same as (x − 3y)(3x + y) = 0. This gives
x = 3y or y = −3x. So all points (3t, t), (t, −3t) are critical points.

For boundary points with g(x, y) = x2 + y 2 = 9, we use the method of Lagrange multiplier
to optimize h(x, y) = 9f (x, y) = 3xy − 4y 2 . Note that ∇g(x, y) = (2x, 2y) ≠ (0, 0) since
x = y = 0 is the only solution but 02 + 02 ≠ 9. Thus, we solve ∇h(x, y) = λ∇g(x, y), i.e.

3y = 2λx,
{
3x − 8y = 2λy.

Combining the equations, we obtain 3y 2 = 2λxy = 3x2 − 8xy. This is again the same as
(x − 3y)(3x + y) = 0, and so the solutions are (3t, t), (t, −3t) (one can solve for t using
x2 + y 2 = 9, but this is not needed).

1 9 1
As f (3t, t) =and f (t, −3t) = − , we see that the global maximum of f is and the
2 2 2
9
global minimum is − .
2

11

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy