QBA 201 Final Exam – 2 hours

Name ____________________________________________ ID no. ____________________

Problem 1. Theoretically, the number of cars sold in a car dealership depends on the market price
of cars and consumers’ income. Using the data for the period from January 2012 to July 2020, the
following OLS estimates are obtained (standard errors are in parentheses):

PRF: Yt = β0 + β1 Pt + β2 Xt + ut

SRF: Yt = b0 + b1 Pt + b2 Xt + et

OLS SRF estimates: Yt = 3.72 – 1.27 Pt + 0.61 Xt

(1.84) (0.451) (0.145) Standard errors

R2 = 0.874, S = 1.235, n = 103

where Yt = the number of cars sold (in units)

Pt = the market price of cars (in hundreds of $)
Xt = consumers’ income (in thousands of $)

a. (1 point) Interpret the coefficient of determination (R2).

b. (3 point) Interpret the OLS point estimate of β0. Interpret the OLS point estimate of β1.
Interpret the OLS point estimate of β2.

c. (2 points) Test whether the market price of cars (Pt) explains the behavior of the number of
cars sold (Yt). In answering, write out the null and alternative hypotheses and the decision
rule. Show your calculations and state your conclusion.

d. (2 points) Test whether consumers’ income (Xt) explains the behavior of the number of cars
sold (Yt). In answering, write out the null and alternative hypotheses and the decision rule.
Show your calculations and state your conclusion.

Page 1 of 14
QBA 201 Final Exam – 2 hours

e. (2 points) Test at the 5% level of significance the null hypothesis that the independent
variables (Pt and Xt) jointly do not explain the behavior of the number of cars sold (Yt). In
answering, write out the null and alternative hypotheses and the decision rule. Show your
calculations and state your conclusion.

f. (2 points) Calculate the 95% confidence interval estimate of σ2. Interpret your result.
It is claimed that the true standard error of regression σ is 1.75. Do you reject this claim?
Clearly explain.

g. (2 points) Explain why we use the OLS method to estimate the unknown population
parameters , , and in the PRF. Include in your answer the condition set by the
Gauss-Marko theorem.

Problem 2. Random sampling from two different populations produced the following results.

̅ = 78 Sx = 32 nx = 101
= 70 Sy = 24 ny = 101

a. (2.5 points) Test the null hypothesis that the variances of the two populations are equal
against a one-sided alternative at the 5% level of significance. In answering, write the null
and alternative hypotheses in addition to the decision rule. Show your calculations and state
your conclusion.

b. (1 point) State clearly the assumption(s) that is necessary for your answer to part (a) to be
valid.

c. (2.5 points) Test the null hypothesis that the population means are equal against a two-sided
alternative. In answering, write the null and alternative hypotheses in addition to the decision
rule. Show your calculations and state your conclusion.

Page 2 of 14
QBA 201 Final Exam – 2 hours

Problem 3. A random sample of 21 AUS students was asked to respond on a scale from one
(strongly disagree) to seven (strongly agree) to the statement: “Sleeping more at night helps
improve students’ grades significantly.” The sample mean response was 5.75 and the sample
standard deviation was .72.

a. (3 points) Test, against a two-sided alternative, the null hypothesis that the population mean
is 5.5. In answering, write the null and alternative hypotheses in addition to the decision
rule. Show your calculations and state your conclusion.

b. (1 point) What assumption(s) is necessary for the test result in part (a) to be valid? Explain
clearly.

c. (1 point) The 2-tails test p-value in part (a) is .127. Given this information, clearly explain
whether you reject the null hypothesis that the population mean is 5.5 against the alternative
that it is greater than 5.5.

Page 3 of 14
QBA 201 Final Exam – 2 hours

Problem 4. Let X1 and X2 be a random sample from a population with mean μ and variance σ2.
Consider the following two point estimators of μ:

= .5 X1 + .5 X2 and ̂ = .3 X1 + .7 X2

a. (2 points) Show that these point estimators are unbiased.

b. (2 points) Which point estimator is a more efficient estimator of μ? Show your work.

c. (1 point) Give an intuitive reason for your answer in part (b).

Page 4 of 14
QBA 201 Final Exam – 2 hours

Formula Sheets

∑ ∑ ̅
= or = …

____________________________________________________________________________________________

Let X be a random variable (and let c be a constant)

Laws of expected value: E(c X) = c E(X) For example, E(10 X ) = 10 E(X)

Laws of variance: V(c X) = c2 V(X) For example, V(10 X) = 100 V(X)

____________________________________________________________________________________________

P(a < Z < b) = P(Z < b) – P(Z< a) P(Z > z) = P(Z < –z) P(Z > –z) = P(Z < z)

P(Z > z) = 1 – P(Z < z)

Page 5
QBA 201 Final Exam – 2 hours

Se( ) = Se( ) = Se( ) =

√ √ √

⁄√
= ~ N(0, 1) ⁄√
= ~ N(0, 1) ⁄√
= ~ N(0, 1+)

– zα/2 < μ< + zα/2 – zα/2 < μ< + zα/2 – tn-1,α/2 < μ< + tn-1,α/2
√ √ √ √ √ √

/
M=
√

/
n=

̅ ̅
| calculated z-value ⁄√
| ≥ critical z-value, z α | calculated z-value ⁄√
| ≥ critical z-value, z α/2

̅ ̅
| calculated z-value ⁄√
| ≥ critical z-value, z α | calculated z-value ⁄√
| ≥ critical z-value, z α/2

̅ ̅
| calculated t-value ⁄√
| ≥ critical t-value, tn-1, α | calculated t-value ⁄√
| ≥ critical t-value, tn-1, α/2

One-tail test critical z-value: 1.28 1.645 2.33

Two-tail test critical z-value: 1.645 1.96 2.58

Page 6
QBA 201 Final Exam – 2 hours

̅‒ ̅‒
| calculated z-value = | ≥ critical z-value, z α | calculated z-value= | ≥ critical z-value, z α/2

____________________________________________________________________________________________

< σ2 < Calculated χ2-value ≥ critical χ2-value, ,

, / , /

, ,α

____________________________________________________________________________________________

^
P= Se ̂ ̂ 1 ̂ /

̂ – zα/2 ̂ 1 ̂ / < p < ̂ + zα/2 ̂ 1 ̂ /

|calculated z-value = |≥ critical z-value, z α |calculated z-value = |≥ critical z-value, z α/2

/ /

Page 7
QBA 201 Final Exam – 2 hours

| Calculated t-value = | | Calculated t-value = | |Calculated t-value = |

One-tail test critical t-value, tn-3, α

Two-tail test critical t-value, tn-3, α/2

Calculated F-value = ≥ critical value, F2,(n-3),α

b0 – tn-3,α/2.Se(b0) < β0 < b0 + tn-3,α/2.Se(b0)

b1 – tn-3,α/2.Se(b1) < β1 < b1 + tn-3,α/2.Se(b1) b2 – tn-3,α/2.Se(b2) < β2 < b2 + tn-3,α/2.Se(b2)

< σ2 <
, ,

Page 8
QBA 201 Final Exam – 2 hours

Page 9
QBA 201 Final Exam – 2 hours

Page 10
QBA 201 Final Exam – 2 hours

Page 11
QBA 201 Final Exam – 2 hours

Page 12
QBA 201 Final Exam – 2 hours

Page 13
QBA 201 Final Exam – 2 hours

Page 14