ISYE 6501 FINAL MODEL REVIEW
TEST QUESTIONS WITH CORRECT
ANSWERS
What is the main objective of K-Means clustering? - Answer-To partition a dataset into a
predefined number of clusters ("k") by grouping data points that are most similar to each
other, minimizing the overall distance between each data point and the centroid of its
assigned cluster
What is the difference between L1 (Lasso) and L2 (Ridge) regularization? - Answer-L1
performs feature selection by shrinking some coefficients to zero, while L2 penalizes all
coefficients equally.
What is the primary purpose of the CUSUM method in time series analysis? - Answer-
To detect small shifts or changes in the mean level of a time series.
How does CUSUM differ from a simple moving average? - Answer-CUSUM monitors
cumulative changes from a target value, while a moving average smooths data over a
window.
What are the key parameters in CUSUM, and what do they control? - Answer-The
reference value (target mean) and decision interval (threshold for detecting changes).
In which scenarios is CUSUM particularly useful? - Answer-For quality control in
manufacturing, detecting shifts in processes, or monitoring financial time series for
anomalies.
What is the key idea behind exponential smoothing in time series forecasting? -
Answer-To assign exponentially decreasing weights to past observations as they get
older.
What are the main types of exponential smoothing? - Answer-Simple Exponential
Smoothing (SES), Holt's Linear Trend Method, and Holt-Winters Seasonal Method.
What is the role of the smoothing parameter alpha in exponential smoothing? - Answer-
It controls how much weight is given to the most recent observation; higher alpha values
give more weight to recent data.
When would you use Holt-Winters smoothing? - Answer-For time series data with both
trend and seasonality.
TEST QUESTIONS WITH CORRECT
ANSWERS
What is the main objective of K-Means clustering? - Answer-To partition a dataset into a
predefined number of clusters ("k") by grouping data points that are most similar to each
other, minimizing the overall distance between each data point and the centroid of its
assigned cluster
What is the difference between L1 (Lasso) and L2 (Ridge) regularization? - Answer-L1
performs feature selection by shrinking some coefficients to zero, while L2 penalizes all
coefficients equally.
What is the primary purpose of the CUSUM method in time series analysis? - Answer-
To detect small shifts or changes in the mean level of a time series.
How does CUSUM differ from a simple moving average? - Answer-CUSUM monitors
cumulative changes from a target value, while a moving average smooths data over a
window.
What are the key parameters in CUSUM, and what do they control? - Answer-The
reference value (target mean) and decision interval (threshold for detecting changes).
In which scenarios is CUSUM particularly useful? - Answer-For quality control in
manufacturing, detecting shifts in processes, or monitoring financial time series for
anomalies.
What is the key idea behind exponential smoothing in time series forecasting? -
Answer-To assign exponentially decreasing weights to past observations as they get
older.
What are the main types of exponential smoothing? - Answer-Simple Exponential
Smoothing (SES), Holt's Linear Trend Method, and Holt-Winters Seasonal Method.
What is the role of the smoothing parameter alpha in exponential smoothing? - Answer-
It controls how much weight is given to the most recent observation; higher alpha values
give more weight to recent data.
When would you use Holt-Winters smoothing? - Answer-For time series data with both
trend and seasonality.
