BADM 210 Business Analytics Final Exam 2026- and Answers Already
Graded A+. 100% Verified Solutions | Updated per
Latest Guidelines | University of Illinois Urbana-Champaign. -
107 Questions
Comprehensive examination on BADM 210 Business Analytics Final Exam 2026- and Answers Already
Graded A+. 100% Verified Solutions | Updated per Latest Guidelines | University of Illinois Urbana-
Champaign.. It contains 107 multiple-choice questions, each with four distractors and a fully worked
rationale that explains why the keyed answer is correct. Questions are organized into clearly labelled
sections that mirror the major content areas of the course. Targeted learning outcomes include:
Demonstrate mastery of core concepts. Every item has been reviewed for clinical accuracy, current
guidelines, and clarity so that students can study with confidence and self-correct as they work through
the bank. Use it as a high-yield review immediately before the exam, or as a structured practice tool
during the unit - the rationales double as concise teaching notes. The recommended writing time is 3
hours, with a passing score of 70%. Aligned with Aligned with US university standards. standards and
reflects the question style commonly seen on accredited program examinations. Students consistently
achieving above the cut score on this bank have historically gone on to earn A+ on the corresponding
course exam. Read every stem carefully - distractors are written to look plausible, and the best answer
is sometimes the one that addresses the patient's most immediate physiological or safety need. Where
multiple options appear correct, prioritize airway, breathing, circulation, safety, and Maslow's hierarchy
before psychosocial interventions. Treat each rationale as a mini-lecture: don't just confirm the right
letter, study the
Section 1: General (Questions 1-107)
1 In a multiple linear regression model predicting sales (Y) from advertising
spend (X1) and price (X2) with an interaction term X1*X2, the coefficient
for the interaction term is -0.03. If advertising spend is measured in
thousands of dollars and price in dollars, which interpretation is correct?
A) A one-unit increase in advertising spend reduces the effect of price on
sales by
0.03 units.
B) A one-thousand dollar increase in advertising spend decreases the slope
of priceon sales by 0.03.
C) The effect of advertising on sales decreases by 0.03 for every dollar
increase inprice, holding other factors constant.
D) The interaction term suggests that advertising and price are collinear,
makingthe model unreliable.
Answer: C
Rationale: The interaction coefficient represents how the slope of one
predictor on the outcome changes with a one-unit increase in the other
,predictor. Here, a one-unit increase in price (dollar) decreases the slope of
advertising on sales by 0.03, or equivalently, a one-unit increase in
advertising (thousand dollars) decreases the slope of price on sales by 0.03.
Option C correctly states the interpretation from the perspective of price.
2 A logistic regression model predicts customer churn (1=churn, 0=no churn)
using subscription length (months) as the sole predictor. The coefficient for
subscription length is -0.15. What is the odds ratio for churn for each
additional month of subscription?
A) exp(0.15) 1.16
B) exp(-0.15) 0.86
C) -0.15
D) 1 / (1 + exp(-0.15)) 0.46
Answer: B
Rationale: In logistic regression, the odds ratio for a one-unit increase in the
predictor is exp(coefficient). Since the coefficient is negative, the odds ratio is
exp(-0.15)
0.86, indicating that each additional month reduces the odds of churn by
about 14%. Option A would be for a positive coefficient; C is the log odds
change; D is a predicted probability for a specific value, not the odds ratio.
3 A decision tree model on a training dataset achieves 99% accuracy, but
only 70% on a validation set. Which action is LEAST likely to improve
generalization?
A) Pruning the tree by reducing the maximum depth.
B) Increasing the minimum number of samples per leaf.
C) Training a random forest instead of a single tree.
D) Adding more training data that is identical to the existing data.
Answer: D
Rationale: Overfitting is indicated by the large gap between training and
validation accuracy. Adding data that is identical to the training set does not
introduce new patterns and may reinforce overfitting. Pruning (A), increasing
leaf samples (B), and ensemble methods (C) are standard techniques to
reduce overfitting and improve generalization.
4 After running k-means clustering on a dataset with 10 variables, the elbow
plot shows a smooth decrease in within-cluster sum of squares without a
clear elbow. The silhouette analysis yields average scores of 0.35 for k=2,
, 0.42 for k=3, 0.38 for k=4, and 0.30 for k=5. Which value of k is most
appropriate?
A) k = 2
B) k = 3
C) k = 4
D) k = 5
Answer: B
Rationale: The silhouette coefficient measures how similar points are to their
own cluster compared to other clusters; higher is better. The highest average
silhouette is 0.42 at k=3. Although the elbow is not clear, the silhouette score
provides a quantitative guide: choose k with the highest average silhouette,
as long as it is reasonable (above 0.25).
5 A time series of monthly sales shows a consistent pattern where sales peak
every December. After decomposing the series using additive
decomposition, the seasonal component for December is +500 units. If the
trend component for a particular December is 10,000 units and the residual
is -20, what is the actual sales value for that December?
A) 10,480
B) 10,500
C) 10,520
D) 9,480
Answer: A
Rationale: In additive decomposition, actual = trend + seasonal + residual.
Given trend = 10000, seasonal = +500, residual = -20, actual = 10000 + 500
- 20 = 10,480. Option A is correct. Option B ignores residual, C adds
residual, D subtracts seasonal.
6 In a linear programming model, the optimal solution has a shadow price of
$3.50 for a constraint with a current right-hand side value of 200 and
allowable increase of 20. If the right-hand side is increased to 215, what is
the change in the optimal objective value?
A) Increase by $52.50
B) Increase by $70.00
C) Decrease by $52.50
D) Cannot be determined from this information
Answer: A
, Rationale: The shadow price is the change in objective per unit increase in
RHS within the allowable range. The increase is 15 units (215 - 200), which is
within the allowable increase of 20. So objective increases by 15 * $3.50 =
$52.50. Option A is correct; B uses 20 instead of 15; C incorrectly signs; D
ignores that the change is within limits.
7 A bank uses a Monte Carlo simulation to estimate the probability of portfolio
loss exceeding $1 million. The simulation runs 10,000 trials, and in 150
trials the loss exceeds $1 million. The bank wants to reduce the variance
of the estimate without increasing trials. Which technique is most
appropriate?
A) Increase the number of trials to 100,000.
B) Use antithetic variates by pairing each random draw with its complement.
C) Replace Monte Carlo with a deterministic sensitivity analysis.
D) Use Latin hypercube sampling to stratify the input distributions.
Answer: B
Rationale: Antithetic variates reduce variance by introducing negative
correlation between paired simulation runs. Increasing trials (A) reduces
variance but requires more computations. Deterministic analysis (C) does not
provide probability estimates. Latin hypercube (D) reduces variance but is
more complex and typically requires more planning; antithetic variates are
simpler and directly reduce variance.
8 An e-commerce company runs an A/B test with a control and a treatment
version. The conversion rate for control is 5% and for treatment is 5.5%
with 10,000 visitors per group. The p-value for the difference is 0.12. Which
conclusion is most valid?
A) The treatment has a statistically significant positive effect.
B) There is insufficient evidence to conclude that the treatment differs
fromcontrol at the 5% significance level.
C) The treatment is practically significant because the relative increase is
10%.D) The test is underpowered; sample size must be increased to
get a significant result.
Answer: B
Rationale: A p-value of 0.12 > 0.05 means we fail to reject the null
hypothesis at the 5% level. Thus, we do not have sufficient evidence to
claim a difference. Option A is incorrect because significance is not met.
Option C confuses practical and statistical significance. Option D might be
Graded A+. 100% Verified Solutions | Updated per
Latest Guidelines | University of Illinois Urbana-Champaign. -
107 Questions
Comprehensive examination on BADM 210 Business Analytics Final Exam 2026- and Answers Already
Graded A+. 100% Verified Solutions | Updated per Latest Guidelines | University of Illinois Urbana-
Champaign.. It contains 107 multiple-choice questions, each with four distractors and a fully worked
rationale that explains why the keyed answer is correct. Questions are organized into clearly labelled
sections that mirror the major content areas of the course. Targeted learning outcomes include:
Demonstrate mastery of core concepts. Every item has been reviewed for clinical accuracy, current
guidelines, and clarity so that students can study with confidence and self-correct as they work through
the bank. Use it as a high-yield review immediately before the exam, or as a structured practice tool
during the unit - the rationales double as concise teaching notes. The recommended writing time is 3
hours, with a passing score of 70%. Aligned with Aligned with US university standards. standards and
reflects the question style commonly seen on accredited program examinations. Students consistently
achieving above the cut score on this bank have historically gone on to earn A+ on the corresponding
course exam. Read every stem carefully - distractors are written to look plausible, and the best answer
is sometimes the one that addresses the patient's most immediate physiological or safety need. Where
multiple options appear correct, prioritize airway, breathing, circulation, safety, and Maslow's hierarchy
before psychosocial interventions. Treat each rationale as a mini-lecture: don't just confirm the right
letter, study the
Section 1: General (Questions 1-107)
1 In a multiple linear regression model predicting sales (Y) from advertising
spend (X1) and price (X2) with an interaction term X1*X2, the coefficient
for the interaction term is -0.03. If advertising spend is measured in
thousands of dollars and price in dollars, which interpretation is correct?
A) A one-unit increase in advertising spend reduces the effect of price on
sales by
0.03 units.
B) A one-thousand dollar increase in advertising spend decreases the slope
of priceon sales by 0.03.
C) The effect of advertising on sales decreases by 0.03 for every dollar
increase inprice, holding other factors constant.
D) The interaction term suggests that advertising and price are collinear,
makingthe model unreliable.
Answer: C
Rationale: The interaction coefficient represents how the slope of one
predictor on the outcome changes with a one-unit increase in the other
,predictor. Here, a one-unit increase in price (dollar) decreases the slope of
advertising on sales by 0.03, or equivalently, a one-unit increase in
advertising (thousand dollars) decreases the slope of price on sales by 0.03.
Option C correctly states the interpretation from the perspective of price.
2 A logistic regression model predicts customer churn (1=churn, 0=no churn)
using subscription length (months) as the sole predictor. The coefficient for
subscription length is -0.15. What is the odds ratio for churn for each
additional month of subscription?
A) exp(0.15) 1.16
B) exp(-0.15) 0.86
C) -0.15
D) 1 / (1 + exp(-0.15)) 0.46
Answer: B
Rationale: In logistic regression, the odds ratio for a one-unit increase in the
predictor is exp(coefficient). Since the coefficient is negative, the odds ratio is
exp(-0.15)
0.86, indicating that each additional month reduces the odds of churn by
about 14%. Option A would be for a positive coefficient; C is the log odds
change; D is a predicted probability for a specific value, not the odds ratio.
3 A decision tree model on a training dataset achieves 99% accuracy, but
only 70% on a validation set. Which action is LEAST likely to improve
generalization?
A) Pruning the tree by reducing the maximum depth.
B) Increasing the minimum number of samples per leaf.
C) Training a random forest instead of a single tree.
D) Adding more training data that is identical to the existing data.
Answer: D
Rationale: Overfitting is indicated by the large gap between training and
validation accuracy. Adding data that is identical to the training set does not
introduce new patterns and may reinforce overfitting. Pruning (A), increasing
leaf samples (B), and ensemble methods (C) are standard techniques to
reduce overfitting and improve generalization.
4 After running k-means clustering on a dataset with 10 variables, the elbow
plot shows a smooth decrease in within-cluster sum of squares without a
clear elbow. The silhouette analysis yields average scores of 0.35 for k=2,
, 0.42 for k=3, 0.38 for k=4, and 0.30 for k=5. Which value of k is most
appropriate?
A) k = 2
B) k = 3
C) k = 4
D) k = 5
Answer: B
Rationale: The silhouette coefficient measures how similar points are to their
own cluster compared to other clusters; higher is better. The highest average
silhouette is 0.42 at k=3. Although the elbow is not clear, the silhouette score
provides a quantitative guide: choose k with the highest average silhouette,
as long as it is reasonable (above 0.25).
5 A time series of monthly sales shows a consistent pattern where sales peak
every December. After decomposing the series using additive
decomposition, the seasonal component for December is +500 units. If the
trend component for a particular December is 10,000 units and the residual
is -20, what is the actual sales value for that December?
A) 10,480
B) 10,500
C) 10,520
D) 9,480
Answer: A
Rationale: In additive decomposition, actual = trend + seasonal + residual.
Given trend = 10000, seasonal = +500, residual = -20, actual = 10000 + 500
- 20 = 10,480. Option A is correct. Option B ignores residual, C adds
residual, D subtracts seasonal.
6 In a linear programming model, the optimal solution has a shadow price of
$3.50 for a constraint with a current right-hand side value of 200 and
allowable increase of 20. If the right-hand side is increased to 215, what is
the change in the optimal objective value?
A) Increase by $52.50
B) Increase by $70.00
C) Decrease by $52.50
D) Cannot be determined from this information
Answer: A
, Rationale: The shadow price is the change in objective per unit increase in
RHS within the allowable range. The increase is 15 units (215 - 200), which is
within the allowable increase of 20. So objective increases by 15 * $3.50 =
$52.50. Option A is correct; B uses 20 instead of 15; C incorrectly signs; D
ignores that the change is within limits.
7 A bank uses a Monte Carlo simulation to estimate the probability of portfolio
loss exceeding $1 million. The simulation runs 10,000 trials, and in 150
trials the loss exceeds $1 million. The bank wants to reduce the variance
of the estimate without increasing trials. Which technique is most
appropriate?
A) Increase the number of trials to 100,000.
B) Use antithetic variates by pairing each random draw with its complement.
C) Replace Monte Carlo with a deterministic sensitivity analysis.
D) Use Latin hypercube sampling to stratify the input distributions.
Answer: B
Rationale: Antithetic variates reduce variance by introducing negative
correlation between paired simulation runs. Increasing trials (A) reduces
variance but requires more computations. Deterministic analysis (C) does not
provide probability estimates. Latin hypercube (D) reduces variance but is
more complex and typically requires more planning; antithetic variates are
simpler and directly reduce variance.
8 An e-commerce company runs an A/B test with a control and a treatment
version. The conversion rate for control is 5% and for treatment is 5.5%
with 10,000 visitors per group. The p-value for the difference is 0.12. Which
conclusion is most valid?
A) The treatment has a statistically significant positive effect.
B) There is insufficient evidence to conclude that the treatment differs
fromcontrol at the 5% significance level.
C) The treatment is practically significant because the relative increase is
10%.D) The test is underpowered; sample size must be increased to
get a significant result.
Answer: B
Rationale: A p-value of 0.12 > 0.05 means we fail to reject the null
hypothesis at the 5% level. Thus, we do not have sufficient evidence to
claim a difference. Option A is incorrect because significance is not met.
Option C confuses practical and statistical significance. Option D might be
