ISYE 6501 Final Quiz | Latest Update 2026/2027 | 200
Questions and Verified Answers | Georgia Tech Introduction
to Analytics Modeling | A+ Graded
1. Which of the following is a nonparametric classification method?
A. Linear regression
B. Logistic regression
C. kNearest Neighbors (kNN)
D. LASSO
Answer: C
Rationale: kNN is nonparametric because it does not assume a specific functional form; the model is
defined by the data points themselves.
2. For which type of problem is fractional factorial design best suited?
A. Clustering
B. Classification
C. Prediction from timeseries data
D. Experimental design
E. Variable selection
Answer: D
Rationale: Fractional factorial designs are used in experimental design to study the effects of multiple
factors efficiently by testing only a subset of all possible factor combinations.
,3. For which type of problem is GARCH best suited?
A. Classification
B. Clustering
C. Experimental design
D. Prediction from timeseries data
E. Variable selection
Answer: D
Rationale: GARCH (Generalized Autoregressive Conditional Heteroscedasticity) models are specifically
designed for timeseries data to model and forecast changing volatility.
4. For which type of problem is LASSO regression best suited?
A. Clustering
B. Classification
C. Experimental design
D. Prediction from timeseries data
E. Variable selection and/or prediction from feature data
Answer: E
Rationale: LASSO (Least Absolute Shrinkage and Selection Operator) performs variable sel ection and
regularization to improve prediction accuracy and interpretability.
5. For which type of problem is ARIMA best suited?
A. Classification
,B. Clustering
C. Experimental design
D. Prediction from timeseries data
E. Variable selection
Answer: D
Rationale: ARIMA (AutoRegressive Integrated Moving Average) combines autoregression, integration,
and moving average components specifically for analyzing and forecasting timeseries data.
6. For which type of problem is kmeans best suited?
A. Clustering
B. Classification
C. Experimental design
D. Prediction from feature data
E. Prediction from timeseries data
Answer: A
Rationale: kmeans is a clustering algorithm used to partition data into k distinct groups based on
similarity.
7. What is the primary difference between supervised and unsupervised learning?
A. Supervised learning uses labeled data; unsupervised learning uses unlabeled data
B. Supervised learning uses unlabeled data; unsupervised learning uses labeled data
C. Both use labeled data
D. Both use unlabeled data
, Answer: A
Rationale: Supervised learning uses labeled data (inputoutput pairs) to train models, while unsupervised
learning finds patterns in unlabeled data.
8. Which model is most appropriate for binary classification?
A. Linear regression
B. Logistic regression
C. kmeans clustering
D. Principal Component Analysis
Answer: B
Rationale: Logistic regression is specifically designed for binary classification problems where the
outcome is categorical with two classes.
9. What is the purpose of crossvalidation?
A. To increase model complexity
B. To evaluate model performance on unseen data
C. To reduce the number of features
D. To speed up computation
Answer: B
Rationale: Crossvalidation is used to assess how the results of a statistical analysis will generalize to an
independent dataset.
Questions and Verified Answers | Georgia Tech Introduction
to Analytics Modeling | A+ Graded
1. Which of the following is a nonparametric classification method?
A. Linear regression
B. Logistic regression
C. kNearest Neighbors (kNN)
D. LASSO
Answer: C
Rationale: kNN is nonparametric because it does not assume a specific functional form; the model is
defined by the data points themselves.
2. For which type of problem is fractional factorial design best suited?
A. Clustering
B. Classification
C. Prediction from timeseries data
D. Experimental design
E. Variable selection
Answer: D
Rationale: Fractional factorial designs are used in experimental design to study the effects of multiple
factors efficiently by testing only a subset of all possible factor combinations.
,3. For which type of problem is GARCH best suited?
A. Classification
B. Clustering
C. Experimental design
D. Prediction from timeseries data
E. Variable selection
Answer: D
Rationale: GARCH (Generalized Autoregressive Conditional Heteroscedasticity) models are specifically
designed for timeseries data to model and forecast changing volatility.
4. For which type of problem is LASSO regression best suited?
A. Clustering
B. Classification
C. Experimental design
D. Prediction from timeseries data
E. Variable selection and/or prediction from feature data
Answer: E
Rationale: LASSO (Least Absolute Shrinkage and Selection Operator) performs variable sel ection and
regularization to improve prediction accuracy and interpretability.
5. For which type of problem is ARIMA best suited?
A. Classification
,B. Clustering
C. Experimental design
D. Prediction from timeseries data
E. Variable selection
Answer: D
Rationale: ARIMA (AutoRegressive Integrated Moving Average) combines autoregression, integration,
and moving average components specifically for analyzing and forecasting timeseries data.
6. For which type of problem is kmeans best suited?
A. Clustering
B. Classification
C. Experimental design
D. Prediction from feature data
E. Prediction from timeseries data
Answer: A
Rationale: kmeans is a clustering algorithm used to partition data into k distinct groups based on
similarity.
7. What is the primary difference between supervised and unsupervised learning?
A. Supervised learning uses labeled data; unsupervised learning uses unlabeled data
B. Supervised learning uses unlabeled data; unsupervised learning uses labeled data
C. Both use labeled data
D. Both use unlabeled data
, Answer: A
Rationale: Supervised learning uses labeled data (inputoutput pairs) to train models, while unsupervised
learning finds patterns in unlabeled data.
8. Which model is most appropriate for binary classification?
A. Linear regression
B. Logistic regression
C. kmeans clustering
D. Principal Component Analysis
Answer: B
Rationale: Logistic regression is specifically designed for binary classification problems where the
outcome is categorical with two classes.
9. What is the purpose of crossvalidation?
A. To increase model complexity
B. To evaluate model performance on unseen data
C. To reduce the number of features
D. To speed up computation
Answer: B
Rationale: Crossvalidation is used to assess how the results of a statistical analysis will generalize to an
independent dataset.
