BUS3059 Quantitative Business Analysis
Midterm Exam Review (Qns & Ans)
2025
Question 1:
Case Study: A retail chain uses multiple regression analysis to
forecast monthly sales. In the model, a dummy variable indicating
a promotional month has a statistically significant coefficient of
15,000.
Question: What does this coefficient imply?
A. Promotional months are associated with a decrease of 15,000
units in sales
B. Promotional months cause a 15,000-unit increase in sales, on
average
C. The variable is insignificant because its effect is constant
D. The dummy variable should be declared non‐numeric
©2025
, Correct ANS: B. Promotional months cause a 15,000‑unit
increase in sales, on average
Rationale: In multiple regression, a dummy variable with a
coefficient of 15,000 indicates that, holding all else equal, being
in a promotional month increases the predicted sales by 15,000
units relative to non‑promotional months.
---
Question 2:
Case Study: A business analyst tests the impact of a new
advertisement campaign on sales using hypothesis testing. The
campaign’s p‑value is reported as 0.03.
Question: What is the most appropriate interpretation at a 5%
significance level?
A. There is a 3% probability that the null hypothesis is true
B. The advertisement has no effect because p‑value is below the
threshold
C. The results are statistically significant, and the campaign likely
has an effect
D. The null hypothesis must be rejected at the 1% significance
level
©2025
, Correct ANS: C. The results are statistically significant, and
the campaign likely has an effect
Rationale: A p‑value of 0.03 is less than the 5% significance
level, which means the null hypothesis (no effect) is rejected,
implying that the new advertisement campaign likely has a
statistically significant impact on sales.
---
Question 3:
Case Study: An analyst is concerned that adding extra predictors
to a regression model might inflate the model’s R‑squared even if
the predictors are not truly relevant.
Question: Which statistic is most useful to assess the model’s
explanatory power while adjusting for the number of predictors?
A. R‑squared
B. Adjusted R‑squared
C. The F‑statistic
D. The Durbin‑Watson statistic
Correct ANS: B. Adjusted R‑squared
Rationale: Adjusted R‑squared corrects for the number of
predictors in the model, providing a more accurate evaluation of
©2025
, the model’s explanatory power by penalizing the addition of
irrelevant variables.
---
Question 4:
Case Study: A company wishes to forecast quarterly revenue
using historical data, which shows both trend and seasonality. The
analyst opts for an exponential smoothing method that accounts
for these factors.
Question: Which method is most appropriate for this
forecasting scenario?
A. Simple Exponential Smoothing
B. Holt’s Linear Trend Method
C. Holt‑Winters Seasonal Method
D. Moving Average
Correct ANS: C. Holt‑Winters Seasonal Method
Rationale: Holt‑Winters seasonal exponential smoothing is
designed to handle data with both trend and seasonal patterns,
making it the ideal method for forecasting quarterly revenue that
fluctuates seasonally.
©2025
Midterm Exam Review (Qns & Ans)
2025
Question 1:
Case Study: A retail chain uses multiple regression analysis to
forecast monthly sales. In the model, a dummy variable indicating
a promotional month has a statistically significant coefficient of
15,000.
Question: What does this coefficient imply?
A. Promotional months are associated with a decrease of 15,000
units in sales
B. Promotional months cause a 15,000-unit increase in sales, on
average
C. The variable is insignificant because its effect is constant
D. The dummy variable should be declared non‐numeric
©2025
, Correct ANS: B. Promotional months cause a 15,000‑unit
increase in sales, on average
Rationale: In multiple regression, a dummy variable with a
coefficient of 15,000 indicates that, holding all else equal, being
in a promotional month increases the predicted sales by 15,000
units relative to non‑promotional months.
---
Question 2:
Case Study: A business analyst tests the impact of a new
advertisement campaign on sales using hypothesis testing. The
campaign’s p‑value is reported as 0.03.
Question: What is the most appropriate interpretation at a 5%
significance level?
A. There is a 3% probability that the null hypothesis is true
B. The advertisement has no effect because p‑value is below the
threshold
C. The results are statistically significant, and the campaign likely
has an effect
D. The null hypothesis must be rejected at the 1% significance
level
©2025
, Correct ANS: C. The results are statistically significant, and
the campaign likely has an effect
Rationale: A p‑value of 0.03 is less than the 5% significance
level, which means the null hypothesis (no effect) is rejected,
implying that the new advertisement campaign likely has a
statistically significant impact on sales.
---
Question 3:
Case Study: An analyst is concerned that adding extra predictors
to a regression model might inflate the model’s R‑squared even if
the predictors are not truly relevant.
Question: Which statistic is most useful to assess the model’s
explanatory power while adjusting for the number of predictors?
A. R‑squared
B. Adjusted R‑squared
C. The F‑statistic
D. The Durbin‑Watson statistic
Correct ANS: B. Adjusted R‑squared
Rationale: Adjusted R‑squared corrects for the number of
predictors in the model, providing a more accurate evaluation of
©2025
, the model’s explanatory power by penalizing the addition of
irrelevant variables.
---
Question 4:
Case Study: A company wishes to forecast quarterly revenue
using historical data, which shows both trend and seasonality. The
analyst opts for an exponential smoothing method that accounts
for these factors.
Question: Which method is most appropriate for this
forecasting scenario?
A. Simple Exponential Smoothing
B. Holt’s Linear Trend Method
C. Holt‑Winters Seasonal Method
D. Moving Average
Correct ANS: C. Holt‑Winters Seasonal Method
Rationale: Holt‑Winters seasonal exponential smoothing is
designed to handle data with both trend and seasonal patterns,
making it the ideal method for forecasting quarterly revenue that
fluctuates seasonally.
©2025
