MATH-UA 212: Math for Economics 2

Midterm 1–Practice

Name:

This exam is scheduled for 75 minutes, to be done individually, without calculators, notes,
textbooks, and other outside materials.

Show all work to receive full credit, except where specified.

Problem Points
Mark an “X” next to your section

X Section TA Class Time

MC and TF /20
7 Vince Zhu F 8:00-9:15am

8 Vince Zhu F 9:30-10:45am

FR 1 /14
9 Hang Fu F 11:00-12:15pm

10 Hang Fu F 12:30-1:45pm
FR 2 /12
12 Ryan Zhao F 8:00-9:15am

13 Ryan Zhao F 9:30-10:45am

FR 3 /9
14 Bill Bao F 11:00-12:15pm

15 Bill Bao F 12:30-1:45pm

FR 4 /10
I pledge that I have observed the NYU honor code, and
that I have neither given nor received unauthorized as- FR 5 /10
sistance during this examination.
Total /75

Signature:
 Math-UA.212.B

Multiple choice, 2 points each. Clearly indicate your answer. There is no partial credit,
and you do not need to show your work.
ln(−y − 1)
1. Let f (x, y) = √ . Find the domain of f .
xy

(a) {(x, y) : x < 0, y < −1}

(b) {(x, y) : −y − 1 > 0}
(c) {(x, y) : x ≤ 0, y ≤ −1}
(d) {(x, y) : xy > 0}
(e) none of these

2. Let f (x, y) = x2 + 2ey . Which of the following vectors is perpendicular to the level
curve f = 6 at (2, 0)?

(a) h4, 0i
(b) h4, −2i
(c) h4, 2i
(d) h4, 1i
(e) none of the above
" #
3 6
3. Let A = . Find A−1 , the inverse of A.
2 5
" #
1 2
(a) A−1 =
2/3 5/3
" #
5/3 −2
(b) A−1 =
−2/3 1
" #
1 −2
(c) A−1 =
−2/3 5/3
" #
1/3 1/6
(d) A−1 =
1/2 1/5
(e) The inverse A−1 does not exist

2
n constraint,
15-3w16 16. f (x, y) = (x + 3)2 + (y − 3)2 , x2 + y 2 ≤ 2
15-3w17 17. f (x, y) = x2 y + 3y 2 − y, x2 + y 2 ≤ 10
15-3w18 18. Decide whether each point appears to be a maximum, Math-UA.212.B
minimum, or neither for the function f constrained by
=5 the loop in Figure 15.30.
4. The
(a) P graph
(b) below
Q shows
(c) level
R sets
(d)of the
S function f (x, y) labeled for f = 10, 20, 30, 40, 50
and 60 and a constraint g(x, y) = c as a loop.

f = 10 Q

Constraint
R
!
=1 P

f = 20
S
f = 50
15-3w18fig f = 30 f = 40 f = 60
!
Which of theFigure
following
15.30 statement(s) is/are true about the constrained optimization
problem for f subject to the constraint g.

I. The point Q appears to be a local minimum of the constrained optimization

II. The points P, Q and S all appear to be a local maximums of the constrained
optimization
III. The point R appears to be neither a max or a min of the constrained optimization
IV. The point S appears to be a global maximum of the constrained optimization

(a) only IV is true

(b) III and IV are true
(c) II, III and IV are true
(d) I, III and IV are true
(e) all statements are true

3
 Math-UA.212.B

5. Initially, three firms A, B, and C share the market for a certain commodity. Firm A
has 35% of the market, Firm B has 25%, and C has 40%. Each year, the following
changes occur:
• A keeps 75% of its customers, while losing 15% to B and 10% to C.
• B keeps 60% of its customers, while losing 5% to A and 35% to C.
• C keeps 65% of its customers, while losing 15% to A and 20% to B.
The share of each company after one year is given by:
 2     
0.75 0.15 0.1 0.35 0.75 0.05 0.15 0.35
(a)  0.05 0.60 0.35   0.25 
   
(c)  0.15 0.6 0.20   0.25 
  
0.15 0.20 0.65 0.4 0.1 0.35 0.65 0.4
  
0.75 0.15 0.1 0.35
(d)  0.05 0.60 0.35   0.25 
    
0.75 0.05 0.15
(b)  0.15 0.6 0.20 
  0.15 0.20 0.65 0.4
0.1 0.35 0.65 (e) None of the above

6. A simple macro model can be described by the three equations:

Y = C + A0
C = a + b(Y − T )
T = d + tY
where Y is income, C is consumption, T is tax revenue, A0 is the constant autonomous
expenditure, and a, b, d, and t are all positive constants. Which matrix equation below
would allow us to find the equilibrium values of the variables Y, C, and T .

1 −b −t Y
    
A0
(a) −1 1 0  C  =  a 
    

0 b 1 T d
      
Y C A0 1
(b) C  =  a  +  b  Y − T 
      

T d t Y

1 −1 0
    
Y A0
(c) −b 1 b  C  =  a 
    

−t 0 1 T d
    
Y C A0
(d) C  =  a  b(Y − T )
    

T d tY

4
 Math-UA.212.B

(e) none of the above

True or false, 2 points each.

If the statement is ever false, circle false as your answer. If the statement is always
true, circle true.
In each case, assume f is a smooth function (its derivatives of all orders exist and are
continuous). If f has a constraint g = c, assume that g is smooth and that ∇g is never
0.

7. If u, v, and w are unit vectors in R2 satisfying u · v = 0 and v · w = 0, then u · w = 0.

TRUE FALSE

1
8. The parallelogram spanned by the vectors i − 4j and i − j has area 2.
2
TRUE FALSE

9. If the function f (x, y) has a minimum at (a, b) subject to the constraint g(x, y) = c,
then f (a, b) ≥ c.
TRUE FALSE

10. If f (x, y) is a differentiable function and (a, b) is a critical point, then the directional
derivative of f must be equal to zero in all directions.
TRUE FALSE

5
 Math-UA.212.B

Free response. You must show all necessary work in order to receive full credit. A correct
answer without supporting work will receive little or no credit.

1. Let Let f (x, y) = xexy − y 2 .

(a) (6 points) Find the directional deriviative of f at (0, −1) in the direction from
(0, −1) to (3, 1).
(b) (5 points) Find a unit vector u such that the directional derivative of f in the
direction of u at (0, −1) is zero. Show all your work to justify your answer!
(c) (3 points) Find the maximum instantaneous rate of change of f at (0, −1).
In other words, find the maximum possible value for Du f (0, −1) if u is a unit
vector.

Solution: (a) We want to compute Du f = ∇f · u where u is a unit vector. To find

∇f at the point (0, −1) we compute

∇f = hexy + xyexy , x2 exy − 2yi = h1, 2i.

(0,−1) (0,−1)

To find u we compute the √ displacement h3 − 0, 1 − (−1)i = h3, 2i and divide by

√ vector √
it’s magnitude |h3, 2i| = 3 + 2 = 9 + 4 = 13. Thus, u = √113 h3, 2i and therefore
2 2

1 7
Du f = ∇f · u = h1, 2i · √ h3, 2i = √
13 13

(b) We want to find a u such that Du f = ∇f · u = 0. Since ∇f h1, 2i, that means
(0,−1)
we want to find a u such that h1, 2i · u = 0. Note that h−2, 1i · h1, 2i = 0. To make
this h−2, 1i a unit vector we divide by its magnitude. That is, take
1 −2 1
u = √ h−2, 1i = h √ , √ i
5 5 5

(c) The maximum instantaneous rate of change at (0, −1) will be |Du f (0, −1)|. There-
fore, the maximum instantenous rate of change of f at (0, −1) is
√
|Du f (0, −1)| = |∇f (0, −1)| = |h1, 2i| = 5

6
 Math-UA.212.B

2. The supply and demand equations for three commodities are:

q1S = 7 + p3 + 5p1 , q1D = 2p2 + 2p1

q2S = −p1 + p2 + 4p3 − 3, q2D = 10 − 2p1

q3S = 3p1 + 4p2 − 4, q3D = p1 + p2 + p3 + 20
 
p1
(a) (6 points) Set up a matrix equation of the form A  p2  = b which would let us
 

p3
solve for the equilibrium prices p1 , p2 , and p3 . That is, find A and b.
You do NOT need to solve this system of equations.
(b) (6 points) Use Cramer’s rule or Gaussian elimination to find p1 , p2 and p3 .

Solution (a) To find equilibrium prices, begin by setting supply and demand
equal to each other (and simplifying)

3p1 − 2p2 + p3 = 7
p1 + p2 + 4p3 = 13
2p1 + 3p2 − p3 = 24

Thus, writing this in matrix form we find

3 −2 1 −7
   

A = 1 1 4  , b =  13 
   

2 3 −1 24

(b) We will show how to use Cramer’s rule to solve for p1 . Solving for p2 , p3 is
similar. Here

|A| = 3(−1 − 12) + 2(−1 − 8) + 1(3 − 2) = · · · = −56

and
−7 −2 1
|D1 | = 13 1 4 = · · · = −112
24 3 −1
−112
Therefore, p1 = = 2.
−56

7
 Math-UA.212.B

3. Given x units of capital and y units of labor, a company produces

U (x, y) = 12x1/3 y 2/3 units of its product.

One unit of capital costs $5 and one unit of labor costs $3.

(a) (6 points) Suppose the company’s budget is $135. Use Lagrange multipliers to
find the values x∗ and y ∗ which maximize production. You may assume that the
max occurs where the Lagrange multipliers condition is satisfied.

Solution We compute

U (x, y) = 12x1/3 y 2/3 , g(x, y) = 5x + 3y = 135

∇U = h4x−2/3 y 2/3 , 8x1/3 y −1/3 i ∇g = h5, 3i

Solving

4x−2/3 y 2/3 = 5λ
8x1/3 y −1/3 = 3λ
5x + 3y = 135

we find x∗ = 9, y ∗ = 30.
(b) (3 points) Without re-doing the optimization problem in part (a); i.e. without
solving for x and y again, estimate the change in the value function U ∗ = U (x∗ , y ∗ )
if the budget changes from $135 to $134. Does U ∗ increase or decrease and by
how much? (you do not need to convert your answer into a decimal)

Solution This question can be answered using the definition of the Langrange
multiplier λ. Namely, note that
 2/3
4 4 4 10
λ = x−2/3 y 2/3 = (9)−2/3 (30)2/3 = · · · =
5 5 5 3

Therefore, if the budget changes from $135 to $134 then the optimum U ∗ will

2/3
decrease by 54 10
3
units.

8
 Math-UA.212.B

4. (10 points) Each week, a charity goes to the market to buy apples, blueberries, and
clementines. The table below lists the number of boxes of apples, blueberries, and
clementines that the charity purchased each week.

Week Apples Blueberries Clementines Total Cost ($)

1 20 30 30 340
2 20 20 40 360
3 10 50 30 400

Let a, b, and c be the cost of a box of apples, blueberries, and clementines, respectively.
Write an augmented matrix to represent the table above, and use Gaussian elimination
to solve for a, b, and c. Explain what row operation you use in each step.
(Note: your final answers should be nice integers.)
Soluton From the information in the chart, we get the following system

20a + 30b + 30c = 240

20a + 20b + 40c = 360
10a + 50b + 30c = 400

which yields the augmented matrix

 
2 3 3 34
 2 2 4 36 
 

1 5 3 40

By Gaussian elimination we find, a = 2, b = 4, c = 6.

9
 Math-UA.212.B

5. (10 points) Consider the following optimization problem

2
max 2 − ey − (x − 1)2 subject to the contraint x2 + y 2 ≤ 1/9.

Find the global maximum of the constrained problem. At which point does the global
maximum occur?
Solution To solve this using the Lagrange multplier method we use the complemen-
tary slackness conditon. That is, we solve ∇f = λ∇g in two scenarios:
1) when λ = 0 and x2 + y 2 < 1/9
2) when λ ≥ 0 and x2 + y 2 = 1/9.

First, we focus on when λ = 0 for when x2 + y 2 < 1/9. That is, we solve ∇f = h−2(x −
2 2
1), −2yey i = 0. Therefore, −2(x − 1) = 0 =⇒ x = 1 and −2yey = 0 =⇒ y = 0.
However, the point (1, 0) is not in the domain of the function, so we discard it.
Next, we focus on the case when λ ≥ 0 for when x2 + y 2 = 1/9. We want to solve
∇f = λ∇g

−2(x − 1) = λ2x
2
−2yey = λ2y
x2 + y 2 = 1/9
2
The second equation implies that y = 0 since if y 6= 0 then we would have λ = −ey < 0
contradicts the assumption that λ ≥ 0. Therefore, knowing y = 0 implies that
which p
x = ± 1/9 = ±1/3. Thus, the solutions are (1/3, 0) and (−1/3, 0).
Plugging these values back into the equation we find,

f (1/3, 0) = · · · = 5/9
f (−1/3, 0) = · · · = −7/9

Therefore, the maximum occurs at (1/3, 0).

10
 Math-UA.212.B

Scrap paper. This page is intentionally left blank for computations!!!

11