ISYE 6501 Introduction to Analytics Modeling Final Quiz Actual Exam 2026/2027 –
Comprehensive Assessment with Detailed Rationales | 100% Verified | Pass
Guaranteed – A+ Graded
Section A: Foundational Models – Regression, Classification, &
Clustering (8 questions)
Q1: A manufacturing analyst fits a multiple linear regression model to predict product
defect rate based on temperature, pressure, and machine age. The residual plot shows
a clear U-shaped pattern. Which assumption of linear regression is most likely violated?
A. Independence of errors
B. Homoscedasticity
C. Linearity [CORRECT]
D. Normality of residuals
Correct Answer: C
Rationale: A U-shaped pattern in residuals indicates a systematic nonlinear relationship
between predictors and the response that the linear model fails to capture; this violates
the linearity assumption, suggesting the need for polynomial terms, transformations, or
a nonlinear model, whereas homoscedasticity violations show funnel shapes and
normality violations show skewed histograms.
Q2: A healthcare analytics team uses logistic regression to predict patient readmission
within 30 days. The model outputs a coefficient of 0.85 for "number of prior
hospitalizations." How should this coefficient be interpreted?
A. Each additional prior hospitalization increases the probability of readmission by 85%
, B. Each additional prior hospitalization increases the odds of readmission by a factor of
e^0.85 ≈ 2.34 [CORRECT]
C. Each additional prior hospitalization decreases the log-odds by 0.85
D. The model predicts readmission with 85% accuracy for patients with prior
hospitalizations
Correct Answer: B
Rationale: In logistic regression, coefficients represent changes in log-odds;
exponentiating the coefficient (e^0.85 ≈ 2.34) gives the odds ratio, meaning each
additional prior hospitalization multiplies the odds of readmission by 2.34, not the
probability, and the sign is positive (increasing, not decreasing) log-odds.
Q3: A retail company applies K-Means clustering to segment 10,000 customers based
on purchase frequency and average order value. The analyst runs the algorithm with K =
3, 5, and 8. Which metric is most appropriate for determining the optimal number of
clusters?
A. Silhouette score [CORRECT]
B. R-squared
C. Mean Absolute Error (MAE)
D. Area Under the ROC Curve (AUC)
Correct Answer: A
Rationale: The silhouette score measures how similar an object is to its own cluster
(cohesion) compared to other clusters (separation), ranging from -1 to +1; it is
specifically designed for clustering evaluation, whereas R-squared and MAE are for
regression, and AUC is for classification performance.
Q4: An analyst performs hierarchical clustering on 50 supply chain facilities using
Euclidean distance. At a certain linkage height, the dendrogram shows two major
branches separating high-volume distribution centers from smaller regional
warehouses. Which linkage method most likely produced this clean separation?
Comprehensive Assessment with Detailed Rationales | 100% Verified | Pass
Guaranteed – A+ Graded
Section A: Foundational Models – Regression, Classification, &
Clustering (8 questions)
Q1: A manufacturing analyst fits a multiple linear regression model to predict product
defect rate based on temperature, pressure, and machine age. The residual plot shows
a clear U-shaped pattern. Which assumption of linear regression is most likely violated?
A. Independence of errors
B. Homoscedasticity
C. Linearity [CORRECT]
D. Normality of residuals
Correct Answer: C
Rationale: A U-shaped pattern in residuals indicates a systematic nonlinear relationship
between predictors and the response that the linear model fails to capture; this violates
the linearity assumption, suggesting the need for polynomial terms, transformations, or
a nonlinear model, whereas homoscedasticity violations show funnel shapes and
normality violations show skewed histograms.
Q2: A healthcare analytics team uses logistic regression to predict patient readmission
within 30 days. The model outputs a coefficient of 0.85 for "number of prior
hospitalizations." How should this coefficient be interpreted?
A. Each additional prior hospitalization increases the probability of readmission by 85%
, B. Each additional prior hospitalization increases the odds of readmission by a factor of
e^0.85 ≈ 2.34 [CORRECT]
C. Each additional prior hospitalization decreases the log-odds by 0.85
D. The model predicts readmission with 85% accuracy for patients with prior
hospitalizations
Correct Answer: B
Rationale: In logistic regression, coefficients represent changes in log-odds;
exponentiating the coefficient (e^0.85 ≈ 2.34) gives the odds ratio, meaning each
additional prior hospitalization multiplies the odds of readmission by 2.34, not the
probability, and the sign is positive (increasing, not decreasing) log-odds.
Q3: A retail company applies K-Means clustering to segment 10,000 customers based
on purchase frequency and average order value. The analyst runs the algorithm with K =
3, 5, and 8. Which metric is most appropriate for determining the optimal number of
clusters?
A. Silhouette score [CORRECT]
B. R-squared
C. Mean Absolute Error (MAE)
D. Area Under the ROC Curve (AUC)
Correct Answer: A
Rationale: The silhouette score measures how similar an object is to its own cluster
(cohesion) compared to other clusters (separation), ranging from -1 to +1; it is
specifically designed for clustering evaluation, whereas R-squared and MAE are for
regression, and AUC is for classification performance.
Q4: An analyst performs hierarchical clustering on 50 supply chain facilities using
Euclidean distance. At a certain linkage height, the dendrogram shows two major
branches separating high-volume distribution centers from smaller regional
warehouses. Which linkage method most likely produced this clean separation?
