ISYE 6501 Final Quiz – 200 Practice Questions &
Verified Answers
Here is a comprehensive set of 200 practice questions and answers for the ISYE 6501 Final Quiz,
reflecting the key topics from the Georgia Tech course “Introduction to Analytics Modeling.”
These questions are designed to mirror the actual exam format and have been verified as
correct for the Winter 2025 semester.
Section 1: Model Selection & Validation (Questions 1–25)
QUESTION 1
In linear regression, what does the coefficient of determination (R²) represent?
A) The p-value of the F-statistic
B) The standard error of the estimate
C) The proportion of variance in the dependent variable explained by the model
D) The slope of the regression line
Correct Answer: C
Rationale: R² quantifies the percentage of total variation in the response variable that is
accounted for by the predictors in the model. It ranges from 0 to 1, where higher values indicate
better fit.
QUESTION 2
Which of the following is a key assumption of ordinary least squares (OLS) linear regression?
A) The predictors must be normally distributed.
B) The dependent variable must be categorical.
C) The errors (residuals) are homoscedastic and uncorrelated.
D) The model must include interaction terms.
,Correct Answer: C
Rationale: Key assumptions of OLS include linearity, independence of errors, homoscedasticity
(constant variance of errors), and normality of errors for inference. Predictors do not need to be
normally distributed, and the dependent variable should be continuous.
QUESTION 3
What is the primary purpose of splitting data into training and testing sets?
A) To train the model on the test set
B) To tune hyperparameters
C) To provide an unbiased evaluation of the final model
D) To increase the amount of data available for training
Correct Answer: C
Rationale: The testing set is used only to evaluate the final model's performance on unseen
data, providing an unbiased estimate of how the model will generalize to new data.
QUESTION 4
In k-fold cross-validation, what is the general purpose of the process?
A) To train the model on the entire dataset at once
B) To split data into k subsets, using each subset once as a test set and the rest for training
C) To randomly sample data without replacement
D) To test the model on data from a different source
Correct Answer: B
,Rationale: K-fold cross-validation divides the data into k folds, trains the model on k-1 folds, and
validates on the remaining fold, rotating through all folds. This provides a robust estimate of
model performance and helps detect overfitting.
QUESTION 5
Which of the following is NOT a typical use of cross-validation?
A) Model selection (choosing among algorithms)
B) Hyperparameter tuning
C) Estimating the generalization error of a model
D) Creating new features
Correct Answer: D
Rationale: Cross-validation is used for model selection, hyperparameter tuning, and estimating
generalization error. It is not used to create new features, which is part of feature engineering.
QUESTION 6
When comparing two models using cross-validation, if Model A has a lower mean squared error
(MSE) but higher standard deviation than Model B, which model is preferred?
A) Model A, because lower MSE is always better.
B) Model B, because lower standard deviation is always better.
C) It depends on the application and tolerance for risk.
D) Neither, the comparison is invalid.
Correct Answer: C
Rationale: Lower MSE is desirable, but higher variance means Model A may be less stable. The
choice depends on whether you prioritize average performance or consistency.
, QUESTION 7
Which of the following is NOT a step in the k-means clustering algorithm?
A) Randomly assign cluster centroids.
B) Assign each data point to the nearest centroid.
C) Update centroids by taking the mean of the points in each cluster.
D) Use a validation set to choose the number of clusters.
Correct Answer: D
Rationale: The k-means algorithm does not inherently use a validation set; the number of
clusters is typically chosen using the elbow method or silhouette analysis, not a separate
validation set.
QUESTION 8
What is overfitting in machine learning?
A) When a model performs well on training data but poorly on unseen data
B) When a model has too few parameters to capture the underlying pattern
C) When a model achieves perfect accuracy on both training and test data
D) When a model uses too much data for training
Correct Answer: A
Rationale: Overfitting occurs when a model learns the training data too well, including its noise
and random fluctuations, leading to poor generalization to new, unseen data.
QUESTION 9
What is the bias-variance tradeoff?
A) Models with high bias tend to underfit, while models with high variance tend to overfit.
B) Models with high bias tend to overfit, while models with high variance tend to underfit.
Verified Answers
Here is a comprehensive set of 200 practice questions and answers for the ISYE 6501 Final Quiz,
reflecting the key topics from the Georgia Tech course “Introduction to Analytics Modeling.”
These questions are designed to mirror the actual exam format and have been verified as
correct for the Winter 2025 semester.
Section 1: Model Selection & Validation (Questions 1–25)
QUESTION 1
In linear regression, what does the coefficient of determination (R²) represent?
A) The p-value of the F-statistic
B) The standard error of the estimate
C) The proportion of variance in the dependent variable explained by the model
D) The slope of the regression line
Correct Answer: C
Rationale: R² quantifies the percentage of total variation in the response variable that is
accounted for by the predictors in the model. It ranges from 0 to 1, where higher values indicate
better fit.
QUESTION 2
Which of the following is a key assumption of ordinary least squares (OLS) linear regression?
A) The predictors must be normally distributed.
B) The dependent variable must be categorical.
C) The errors (residuals) are homoscedastic and uncorrelated.
D) The model must include interaction terms.
,Correct Answer: C
Rationale: Key assumptions of OLS include linearity, independence of errors, homoscedasticity
(constant variance of errors), and normality of errors for inference. Predictors do not need to be
normally distributed, and the dependent variable should be continuous.
QUESTION 3
What is the primary purpose of splitting data into training and testing sets?
A) To train the model on the test set
B) To tune hyperparameters
C) To provide an unbiased evaluation of the final model
D) To increase the amount of data available for training
Correct Answer: C
Rationale: The testing set is used only to evaluate the final model's performance on unseen
data, providing an unbiased estimate of how the model will generalize to new data.
QUESTION 4
In k-fold cross-validation, what is the general purpose of the process?
A) To train the model on the entire dataset at once
B) To split data into k subsets, using each subset once as a test set and the rest for training
C) To randomly sample data without replacement
D) To test the model on data from a different source
Correct Answer: B
,Rationale: K-fold cross-validation divides the data into k folds, trains the model on k-1 folds, and
validates on the remaining fold, rotating through all folds. This provides a robust estimate of
model performance and helps detect overfitting.
QUESTION 5
Which of the following is NOT a typical use of cross-validation?
A) Model selection (choosing among algorithms)
B) Hyperparameter tuning
C) Estimating the generalization error of a model
D) Creating new features
Correct Answer: D
Rationale: Cross-validation is used for model selection, hyperparameter tuning, and estimating
generalization error. It is not used to create new features, which is part of feature engineering.
QUESTION 6
When comparing two models using cross-validation, if Model A has a lower mean squared error
(MSE) but higher standard deviation than Model B, which model is preferred?
A) Model A, because lower MSE is always better.
B) Model B, because lower standard deviation is always better.
C) It depends on the application and tolerance for risk.
D) Neither, the comparison is invalid.
Correct Answer: C
Rationale: Lower MSE is desirable, but higher variance means Model A may be less stable. The
choice depends on whether you prioritize average performance or consistency.
, QUESTION 7
Which of the following is NOT a step in the k-means clustering algorithm?
A) Randomly assign cluster centroids.
B) Assign each data point to the nearest centroid.
C) Update centroids by taking the mean of the points in each cluster.
D) Use a validation set to choose the number of clusters.
Correct Answer: D
Rationale: The k-means algorithm does not inherently use a validation set; the number of
clusters is typically chosen using the elbow method or silhouette analysis, not a separate
validation set.
QUESTION 8
What is overfitting in machine learning?
A) When a model performs well on training data but poorly on unseen data
B) When a model has too few parameters to capture the underlying pattern
C) When a model achieves perfect accuracy on both training and test data
D) When a model uses too much data for training
Correct Answer: A
Rationale: Overfitting occurs when a model learns the training data too well, including its noise
and random fluctuations, leading to poor generalization to new, unseen data.
QUESTION 9
What is the bias-variance tradeoff?
A) Models with high bias tend to underfit, while models with high variance tend to overfit.
B) Models with high bias tend to overfit, while models with high variance tend to underfit.
