ISYE 6402 MIDTERM-2 EXAM |
QUESTIONS AND FULL SOLUTIONS |
2026 UPDATED | 100% CORRECT - GT.
116 Questions with Answers and Detailed Rationales
100 PERCENT GUARANTEED PASS
INSTANT DOWNLOAD ANSWERS INCLUDED
IMPORTANCE OF THIS DOCUMENT
This comprehensive examination preparation guide has been meticulously developed to help you succeed in the
ISYE 6402 MIDTERM-2 EXAM | QUESTIONS AND FULL SOLUTIONS | 2026 UPDATED | 100% CORRECT -
GT.. It contains 116 carefully selected questions that reflect the most current exam content and testing strategies.
Each question is accompanied by a correct answer and a detailed rationale that explains the underlying
pathophysiology, pharmacology, or clinical reasoning.
Self-Assessment – Test your knowledge and Exam Preparation – Familiarize yourself with the
identify areas requiring further question format and content
study areas
Concept Reinforcement – Deepen your Confidence Building – Develop test-taking
understanding through strategies and reduce
evidence-based exam anxiety
rationales
Time Management – Practice answering
questions under simulated
exam conditions
Review Summary 116 Questions
Foundations - Application - ISYE 6402 -2 AND FULL Solutions 2026 Updated 100 Correct - GT TIME
Series Analysis AND Forecasting Graduate
All answers with rationales
,Table of Contents
Content Area Questions Key Topics
TIME Series Regression AND 1-20 Model, Series, Arima, Seasonal, Analysis
Forecasting
Arima Models 21-40 Model, Series, Arima, Function, Trend
Seasonality AND 41-60 Process, Chain, State, Markov, Variance
Decomposition
Model Selection AND 61-80 Model, Series, Process, Poisson, Function
Diagnostics
Exponential Smoothing 81-100 Model, TIME Series, Seasonal, Arima, Context
Stationarity AND Differencing 101-116 Solution, Function, Stochastic, Recourse, Problem
TOTAL 116 All questions include answers and detailed rationales
,Section A - TIME Series Regression AND Forecasting
Q1.
Given a non-stationary time series with a unit root, which transformation is most
appropriate to achieve stationarity before applying ARMA modeling?
A. Logarithmic transformation B. First differencing
C. Square root transformation D. Box-Cox transformation
Correct: B - First differencing
Rationale:First differencing removes a unit root, making the series stationary. Logarithmic or
Box-Cox transformations stabilize variance but do not address non-stationarity in the mean.
Square root is also variance-stabilizing.
Q2.
In spectral analysis, what does the peak in the periodogram indicate?
A. The dominant frequency of the series B. The presence of a stochastic trend
C. The order of the AR process D. The variance of the white noise
Correct: A - The dominant frequency of the series
Rationale:The periodogram estimates the spectral density; a peak indicates a dominant
cyclical component at that frequency. Trends appear at low frequencies, not as peaks. AR
order and noise variance are not directly indicated by peaks.
Q3.
When comparing a state-space model to a traditional ARIMA model, which advantage
does the state-space formulation provide?
A. It eliminates the need for parameter B. It can handle missing observations and
estimation time-varying parameters
C. It always produces smaller forecast D. It requires no distributional assumptions
intervals on errors
Correct: B - It can handle missing observations and time-varying parameters
Rationale:State-space models naturally accommodate missing data and allow parameters to
evolve over time. They still require estimation and distributional assumptions. Forecast
intervals are not universally smaller.
Page 3
, Section A - TIME Series Regression AND Forecasting
Q4.
You fit an AR(2) model to a series and compute the ACF of residuals. Several lags exceed
the 95% confidence bounds. What is the most appropriate next step?
A. Accept the model because the ACF is B. Increase the AR order to AR(3) or AR(4)
only a diagnostic
C. Use a MA(1) model instead D. Apply a nonlinear transformation to the
data
Correct: B - Increase the AR order to AR(3) or AR(4)
Rationale:Significant residual autocorrelation indicates model inadequacy. Increasing the AR
order can capture remaining autocorrelation. Switching to MA or transforming may not
address the issue directly.
Q5.
Which model selection criterion penalizes model complexity most heavily for large sample
sizes?
A. AIC B. BIC
C. AICc D. Adjusted R-squared
Correct: B - BIC
Rationale:BIC has a stronger penalty term (log(n) * k) than AIC (2k), making it more
conservative for large n. AICc is a correction for small samples. Adjusted R-squared is not a
model selection criterion.
Q6.
In the Kalman filter, what does the innovation represent?
A. The difference between the predicted and B. The estimated state vector
actual observation
C. The covariance of the state error D. The measurement noise variance
Correct: A - The difference between the predicted and actual observation
Rationale:The innovation is the new information from the observation, defined as y_t -
H*x_t|t-1. It drives the state update. The other options are components of the filter but not the
innovation.
Page 4
QUESTIONS AND FULL SOLUTIONS |
2026 UPDATED | 100% CORRECT - GT.
116 Questions with Answers and Detailed Rationales
100 PERCENT GUARANTEED PASS
INSTANT DOWNLOAD ANSWERS INCLUDED
IMPORTANCE OF THIS DOCUMENT
This comprehensive examination preparation guide has been meticulously developed to help you succeed in the
ISYE 6402 MIDTERM-2 EXAM | QUESTIONS AND FULL SOLUTIONS | 2026 UPDATED | 100% CORRECT -
GT.. It contains 116 carefully selected questions that reflect the most current exam content and testing strategies.
Each question is accompanied by a correct answer and a detailed rationale that explains the underlying
pathophysiology, pharmacology, or clinical reasoning.
Self-Assessment – Test your knowledge and Exam Preparation – Familiarize yourself with the
identify areas requiring further question format and content
study areas
Concept Reinforcement – Deepen your Confidence Building – Develop test-taking
understanding through strategies and reduce
evidence-based exam anxiety
rationales
Time Management – Practice answering
questions under simulated
exam conditions
Review Summary 116 Questions
Foundations - Application - ISYE 6402 -2 AND FULL Solutions 2026 Updated 100 Correct - GT TIME
Series Analysis AND Forecasting Graduate
All answers with rationales
,Table of Contents
Content Area Questions Key Topics
TIME Series Regression AND 1-20 Model, Series, Arima, Seasonal, Analysis
Forecasting
Arima Models 21-40 Model, Series, Arima, Function, Trend
Seasonality AND 41-60 Process, Chain, State, Markov, Variance
Decomposition
Model Selection AND 61-80 Model, Series, Process, Poisson, Function
Diagnostics
Exponential Smoothing 81-100 Model, TIME Series, Seasonal, Arima, Context
Stationarity AND Differencing 101-116 Solution, Function, Stochastic, Recourse, Problem
TOTAL 116 All questions include answers and detailed rationales
,Section A - TIME Series Regression AND Forecasting
Q1.
Given a non-stationary time series with a unit root, which transformation is most
appropriate to achieve stationarity before applying ARMA modeling?
A. Logarithmic transformation B. First differencing
C. Square root transformation D. Box-Cox transformation
Correct: B - First differencing
Rationale:First differencing removes a unit root, making the series stationary. Logarithmic or
Box-Cox transformations stabilize variance but do not address non-stationarity in the mean.
Square root is also variance-stabilizing.
Q2.
In spectral analysis, what does the peak in the periodogram indicate?
A. The dominant frequency of the series B. The presence of a stochastic trend
C. The order of the AR process D. The variance of the white noise
Correct: A - The dominant frequency of the series
Rationale:The periodogram estimates the spectral density; a peak indicates a dominant
cyclical component at that frequency. Trends appear at low frequencies, not as peaks. AR
order and noise variance are not directly indicated by peaks.
Q3.
When comparing a state-space model to a traditional ARIMA model, which advantage
does the state-space formulation provide?
A. It eliminates the need for parameter B. It can handle missing observations and
estimation time-varying parameters
C. It always produces smaller forecast D. It requires no distributional assumptions
intervals on errors
Correct: B - It can handle missing observations and time-varying parameters
Rationale:State-space models naturally accommodate missing data and allow parameters to
evolve over time. They still require estimation and distributional assumptions. Forecast
intervals are not universally smaller.
Page 3
, Section A - TIME Series Regression AND Forecasting
Q4.
You fit an AR(2) model to a series and compute the ACF of residuals. Several lags exceed
the 95% confidence bounds. What is the most appropriate next step?
A. Accept the model because the ACF is B. Increase the AR order to AR(3) or AR(4)
only a diagnostic
C. Use a MA(1) model instead D. Apply a nonlinear transformation to the
data
Correct: B - Increase the AR order to AR(3) or AR(4)
Rationale:Significant residual autocorrelation indicates model inadequacy. Increasing the AR
order can capture remaining autocorrelation. Switching to MA or transforming may not
address the issue directly.
Q5.
Which model selection criterion penalizes model complexity most heavily for large sample
sizes?
A. AIC B. BIC
C. AICc D. Adjusted R-squared
Correct: B - BIC
Rationale:BIC has a stronger penalty term (log(n) * k) than AIC (2k), making it more
conservative for large n. AICc is a correction for small samples. Adjusted R-squared is not a
model selection criterion.
Q6.
In the Kalman filter, what does the innovation represent?
A. The difference between the predicted and B. The estimated state vector
actual observation
C. The covariance of the state error D. The measurement noise variance
Correct: A - The difference between the predicted and actual observation
Rationale:The innovation is the new information from the observation, defined as y_t -
H*x_t|t-1. It drives the state update. The other options are components of the filter but not the
innovation.
Page 4
