Spring 2021 MGT 6203 FINAL EXAM
PART1 – THEORY
(It took me 69 minutes on 5 May 2021)
Week 1
Q1) What would be the null hypothesis for the regression F-test for the following equation
Y = β0 + β1 x1 + β2 x2 + u
A. β1 = 0 and β2 = 0
B. β1 = 0 or β2 = 0
C. β0 = 0 and β1 = 0 and β2 = 0
D. β0 = 0 or β1 = 0 or β2 = 0
Answer: A.
Explanation: the regression F-statistics test the joint null hypothesis that the values of all
the coefficients of explanatory variables except the constant are 0. The intercept is not
included in the hypothesis. (Week 1, slide 24 and 32)
Week 2
Q2) We have the following regression equation:
life_span = B0 + B1*HoursExercisePerWeek + B2*Smoker + B3*Exer_Smoker
HoursExercisePerWeek is a continuous variable
Smoker is a dummy variable (1 is smoker, 0 is non-smoker)
Exer_Smoker is an interaction term HoursExercisePerWeek* Smoker
Which variable impacts the intercept on the y axis?
A. HoursExercisePerWeek (b1)
B. Smoker (b2)
C. Exer_Smoker (b3)
D. All of the above
Answer: B. Week 2, page 19, slide 2
Explanation: HoursExercisePerWeek (b1) and Exer_Smoker (b3) impact the slope only.
Week 3
Q3) As X increases by 1%, y changes by b1% holding all other factors constant. Which model
can be interpreted like this?
, A. Log-Linear Model
B. Log-Log Model
C. Linear-Log Model
D. Polynomial
Answer: B. Week 3, slide 17, upper
Week 4
Q4) A group of 20 students spend between 0 and 6 hours studying for an exam (number of
hours being the predictor). The response variable has 2 outcomes: Pass=1, Fail=0. The
logistic regression analysis gives the following output:
Coefficient Std. Error z-value P-value
Intercept −4.0777 1.7610 −2.316 0.0206
Hours 1.5046 0.6287 2.393 0.0167
What is the probability of passing the exam for a student who studies 5 hours?
A. 0.87
B. 0.96
C. 0.34
D. 0.61
Answer: B. week 4, page 19, s1
Explanation: p= e(b0+b1*x)/(1+ e(b0+b1*x)) = e(-4.0777+5*1.5046)/(1+ e(-4.0777+5*1.5046)) = 0.96
Week 5
Q5). Which of the following is not needed to establish causation?
a) Hypothesized cause must precede its anticipated effect
b) Other possible explanations that can cause the effect must be ruled out
c) Change in cause must lead to a change in effect
d) The effect must always have a reverse impact on the cause
Answer: D. (week 5, page 3)
Week 6
Q6) Using the stock monthly return in the chart below – calculate the Compounded return.
You may use the standalone built-in calculator app on your computer, but advanced
statistical tools like Excel, Python, R are prohibited and will trigger a violation.
Date Price Dividend Stock Split
Jan 45
Feb 26 2 for 1
March 27.2 0.6
April 17 3 for 2
PART1 – THEORY
(It took me 69 minutes on 5 May 2021)
Week 1
Q1) What would be the null hypothesis for the regression F-test for the following equation
Y = β0 + β1 x1 + β2 x2 + u
A. β1 = 0 and β2 = 0
B. β1 = 0 or β2 = 0
C. β0 = 0 and β1 = 0 and β2 = 0
D. β0 = 0 or β1 = 0 or β2 = 0
Answer: A.
Explanation: the regression F-statistics test the joint null hypothesis that the values of all
the coefficients of explanatory variables except the constant are 0. The intercept is not
included in the hypothesis. (Week 1, slide 24 and 32)
Week 2
Q2) We have the following regression equation:
life_span = B0 + B1*HoursExercisePerWeek + B2*Smoker + B3*Exer_Smoker
HoursExercisePerWeek is a continuous variable
Smoker is a dummy variable (1 is smoker, 0 is non-smoker)
Exer_Smoker is an interaction term HoursExercisePerWeek* Smoker
Which variable impacts the intercept on the y axis?
A. HoursExercisePerWeek (b1)
B. Smoker (b2)
C. Exer_Smoker (b3)
D. All of the above
Answer: B. Week 2, page 19, slide 2
Explanation: HoursExercisePerWeek (b1) and Exer_Smoker (b3) impact the slope only.
Week 3
Q3) As X increases by 1%, y changes by b1% holding all other factors constant. Which model
can be interpreted like this?
, A. Log-Linear Model
B. Log-Log Model
C. Linear-Log Model
D. Polynomial
Answer: B. Week 3, slide 17, upper
Week 4
Q4) A group of 20 students spend between 0 and 6 hours studying for an exam (number of
hours being the predictor). The response variable has 2 outcomes: Pass=1, Fail=0. The
logistic regression analysis gives the following output:
Coefficient Std. Error z-value P-value
Intercept −4.0777 1.7610 −2.316 0.0206
Hours 1.5046 0.6287 2.393 0.0167
What is the probability of passing the exam for a student who studies 5 hours?
A. 0.87
B. 0.96
C. 0.34
D. 0.61
Answer: B. week 4, page 19, s1
Explanation: p= e(b0+b1*x)/(1+ e(b0+b1*x)) = e(-4.0777+5*1.5046)/(1+ e(-4.0777+5*1.5046)) = 0.96
Week 5
Q5). Which of the following is not needed to establish causation?
a) Hypothesized cause must precede its anticipated effect
b) Other possible explanations that can cause the effect must be ruled out
c) Change in cause must lead to a change in effect
d) The effect must always have a reverse impact on the cause
Answer: D. (week 5, page 3)
Week 6
Q6) Using the stock monthly return in the chart below – calculate the Compounded return.
You may use the standalone built-in calculator app on your computer, but advanced
statistical tools like Excel, Python, R are prohibited and will trigger a violation.
Date Price Dividend Stock Split
Jan 45
Feb 26 2 for 1
March 27.2 0.6
April 17 3 for 2
