2019 EXAMINATIONS

PART I

ECONOMICS

ECON 103 - Quantitative Methods for Economics Main Paper (3 hours)

Candidates should answer ALL questions and give detailed answers. You can use a non-programmable
calculator.
Good Luck!

1 (Please Turn Over)

SECTION A: Mathematics [50 marks]

1. Let f (x) = x−2/3 + x1/3 for 1.2 ≤ x ≤ 3.5.

(a) Find all stationary points of f .
(b) Find the global minimum and maximum of f .
(c) Find the intervals where f is convex or concave.
(11 marks)
2. A farmer wants to fence a rectangular grazing area along a straight river (no fence is needed along the river).
There are 1700 total feet of fencing available. What dimensions (length and width) will maximize the grazing
area? (7 marks)
Rb Rc Rb b
3. Assume that a f (x)dx = 9, b f (x)dx = −7, a (f (x))2 dx = 36, and a g(x)dx = −2. Use this information
R

to calculate each of the following.

Rc
(a) a f (x)dx
Rb R 2
b
(b) a (f (x))2 dx − a f (x)dx
Rb
(c) a (2f (x) − g(x))dx
(7 marks)

4. A consumer has the following utility function:

U (x, y) = 5x2 + 6xy + y 2 + 38x + 18y

where x and y represent the quantities consumed of two commodities.

(a) Write down the expressions of the marginal utilities of the two commodities. (2 marks)
(b) The consumer has a budget of $40. A unit of the first commodity costs $10, and a unit of the second
$5. The consumer wants to maximize his utility subject to his budget constraint. Using the Lagrange
multiplier method, find the values of x and y which solve the constrained optimization problem of the
consumer. (8 marks)
(c) Compute the value of the Lagrange multiplier λ and provide an interpretation of it. (2 marks)
5. (a) For what values of a does matrix  
2 −1 0
A =  −1 1 1 
0 a 2
have an inverse? (2 marks)
−1
(b) Find the entries in the first row of A , when a = 2. (3 marks)
6. A monopoly produces two goods with the following demand functions:

q1 = 100 − 2p1 + p2
q2 = 120 + 3p1 − 5p2

(a) Write down the expression of the profit function of the firm if its total costs (TC) are given by:

T C = 50 + 10q1 + 20q2

where q1 and q2 are the outputs of the two goods, and p1 and p2 are the prices of the goods. (3 marks)
(b) Use Cramer’s rule to find the values of p1 and p2 which maximize the monopoly’s profit. (5 marks)

(Please Turn Over) 2

SECTION B: Statistics [50 marks]

1. Suppose that 15% of all invoices are for amounts greater than £1,000. A random sample of 60 invoices is
taken.
(a) What is the mean and standard error of the sample proportion of invoices with amounts in excess of
£1,000? (6 marks)
(b) What is the probability that the proportion of invoices in the sample is greater than 18%? (6 marks)
2. Suppose that there exists an imperfect test for cancer. If someone has cancer, there is a 95% chance that the
test will come up ”red”, i.e. positive. If someone does not have cancer, there is only a 2% chance that the
test will come up red. Finally, the chance that anyone has cancer is, say, 5%. Once someone takes the test
and it comes up red, what is the probability that they have cancer? (13 marks)
3. An electronic toothbrush brand claims that at least 80% of dentists recommend their toothbrushes. Suppose
you are an analyst who works for an independent market research company and you want to test the brand’s
claim.

(a) State the appropriate null hypothesis and alternative hypothesis for your test. (3 marks)
(b) Briefly describe how you would conduct the market research to collect data to perform the test.
(4 marks)
(c) Suppose the test statistics you calculated based on your data is z = −2.086. What is your conclusion?
(Note: First choose an appropriate significance level for the test.) (3 marks)

4. Suppose you are interested in the relationship between x and y. You have obtained the following regression
output:

ANOVA
df SS MS F Significance F
Regression 1 13.6693 13.6693 1.8258 0.1925
Residual 19 142.2513 7.4869
Total 20 155.9206

Coefficients Standard Error t Stat P-value

Intercept 12.2691 2.7435 4.4721 0.0003
x -0.3863 0.2859 -1.3512 0.1925

(a) Compute the coefficient of determination and interpret the result. (3 marks)
(b) What is the estimated value of y when x = 10? (3 marks)
(c) Calculate the sample correlation coefficient and explain your result. (3 marks)
(d) Determine whether there is a linear relationship between x and y and explain your conclusion.
(3 marks)
(e) Calculate the sample variation of x: Sx2 . (3 marks)

3 (End of Paper)
1 Lines 5 Differentiation Rules

Slope of line through (x1 , y1 ) and (x2 , y2 ): d α

x =αxα−1
dx
d 1
y2 − y1 ln(x) =
m= dx x
x2 − x1 d x
e =ex
dx
Equation of line through (x1 , y1 ) with slope m: d
cos(x) = − sin(x)
dx
d
y = m(x − x1 ) + y1 sin(x) = cos(x)
dx
d x
a =ax ln a (a > 0)
Equation of line with slope m and y-intercept b: dx
d
(f (x) ± g(x)) =f 0 (x) ± g 0 (x)
dx
y = b + mx
d
kf (x) =kf 0 (x)
dx
d
f (x)g(x) =f 0 (x)g(x) + f (x)g 0 (x)
dx
2 Rules of Exponents d f (x) f 0 (x)g(x) − f (x)g 0 (x)
=
dx g(x) (g(x))2
d
f (g(x)) =f 0 (g(x))g 0 (x)
dx
ax at = ax+t
ax
= ax−t
6 Integration Rules
at
(ax )t = axt Z
1
xn dx = xn+1 + C, n 6= −1
n+1
Z
1
dx = ln |x| + C
x
3 Natural Log Z
1 x
ax dx = a + C, a > 0
ln a
Z
Definition: y = ln x means ey = x. ln xdx = x ln x − x + C
Z
sin xdx = − cos x + C
ln ex = x Z
eln x = x cos xdx = sin x + C
ln(AB) = ln A + ln B
A
ln( ) = ln A − ln B
7 Integration by parts and substi-
B
ln Ap = p ln A
tution
Z Z
0
u(x)v (x)dx = u(x)v(x) − u0 (x)v(x)dx
Z ϕ(b) Z b
4 Quadratic Formula f (x)dx = f (ϕ(y))ϕ0 (y)dy
ϕ(a) a

If ax2 + bx + c = 0, then
8 Inverse of 2× 2 matrix
√  −1  
−b ± b2 − 4ac a b 1 d −b
x= =
2a c d ad − bc −c a

4
9 Statistics for Tests of Hypothe- • Test for Zero Population Correlation:
sis √
r n−2
t= √
• Tests of the Mean of a Normal Population (Case 1) 1 − r2

X̄ − µ0
Z= √ ∼ N (0, 1)
σ/ n 11 Regression Analysis
• Tests of the Mean of a Normal Distribution (Case • Ordinary Least Square Coefficient Estimators:
2)
X̄ − µ0
P
T = √ ∼ tn−1 (xi − x̄)(yi − ȳ)
b1 =
SX / n
P
(xi − x̄)2
• Tests of the Population Proportion (Large Sam-
ples) b0 = ȳ − b1 x̄
p̂ − p0
Z=p ∼ N (0, 1)
p0 (1 − p0 )/n • Estimation of Model Error Variance (σ2 ):
• Tests of the Difference Between Population Means
(yi − ŷi )2
P
(Case 1) Se2 =
n−2
x̄ − ȳ
Z=q ∼ N (0, 1)
2
σx σy2
nx + ny • Unbiased estimator of σb21 :

• Tests of the Difference Between Population Means S2

(Case 2) Sb21 = P e
(xi − x̄)2
x̄ − ȳ
T =q ∼ tnx +ny −2
Sp2 Sp2
+ • Test statistics for hypothesis testing of H0 : β1 = 0:
nx ny

2
(nx −1)Sx +(ny −1)Sy2 b1 − 0
where Sp2 = nx +ny −2 . T = ∼ tn−2
Sb1
• Tests of the Difference Between two Proportion
(Large Samples)
• Confidence intervals: (for the forecast of the con-
p̂x − p̂y ditional expectation E[yn+1 |xn+1 ])
Z=q ∼ N (0, 1)
p̂0 (1−p̂0 ) p̂0 (1−p̂0 )
nx + ny
s
1 (xn+1 − x̄)2
nx p̂x +ny p̂y
ŷn+1 ± tn−2,α/2 Se + P
where p̂0 = n (xi − x̄)2
nx +ny .

10 Correlation • Prediction intervals: (for the forecast of the actual

value resulting for yn+1 )
• Pearson Correlation Coefficient: s
(xn+1 − x̄)2
P
(xi − x̄)(yi − ȳ) 1
r = pP ŷn+1 ± tn−2,α/2 Se 1+ + P
(xi − x̄)2
pP
(xi − x̄)2 (yi − ȳ)2 n

5
6
7