ISYE 6644-ASY, OAN Exam 1 Simulation | Questions and
Answers - 2026 Update | 100% Correct - GT. - 107 Questions and
Answers Already Graded A+ Premium Exam Tested And
Verified
Subject Area ISYE 6644-ASY, OAN Exam 1 Simulation | Questions and Answers - 2026
Update | 100% Correct - GT.
Description Comprehensive examination on ISYE 6644-ASY, OAN Exam 1 Simulation |
Questions and Answers - 2026 Update | 100% Correct - GT..
Expected Grade A+
Total Questions 107
Duration 3 hours
Learning Outcomes 1. Demonstrate mastery of core concepts
Accreditation Aligned with US university standards.
Page 1
,Question 1 of 107
In steady-state simulation output analysis, the batch means method is used to estimate the variance
of the sample mean. If the batch size is too small, the resulting confidence interval may be unreliable
because the batch means are correlated. Which condition is essential for the batch means method to
produce a valid confidence interval?
A. The number of batches must be at least 30 to ensure normality.
B. The batch means must be approximately independent and identically distributed.
C. The initial transient bias must be removed via a warm-up period.
D. The batch size must be equal to the number of replications.
The correct answer is:
Correct Action: The batch means must be approximately independent and identically distributed.
Rationales
• The batch means must be approximately independent and identically distributed. (Correct):
This is the correct action. The batch means method relies on the batch means being approximately i.i.d. If batch size is too
small, autocorrelation among batch means persists, invalidating the standard variance estimator and confidence interval.
Removing initial transient (C) is necessary but not the core condition for the batch means' validity.
• The number of batches must be at least 30 to ensure normality. (Incorrect):
This option is not appropriate. i.d
• The initial transient bias must be removed via a warm-up period. (Incorrect):
This option is not appropriate. i.d
• The batch size must be equal to the number of replications. (Incorrect):
This option is not appropriate. i.d
Page 2
,Question 2 of 107
A linear congruential generator (LCG) has modulus m = 2^31 - 1, multiplier a = 950706376, and
increment c = 0. The spectral test is applied to assess the generator's lattice structure in k
dimensions. Which of the following best interprets a poor spectral test result in 2 dimensions?
A. The generator's period is less than 2^10.
B. Points (u_i, u_{i+1}) fall on a small number of parallel hyperplanes.
C. The generator fails the chi-square goodness-of-fit test.
D. The sequence exhibits a long positive autocorrelation at lag 1.
The correct answer is:
Correct Action: Points (u_i, u_{i+1}) fall on a small number of parallel hyperplanes.
Rationales
• Points (u_i, u_{i+1}) fall on a small number of parallel hyperplanes. (Correct):
This is the correct action. The spectral test measures the maximum distance between parallel hyperplanes covering all
points. A poor result means points lie on few widely spaced hyperplanes, indicating poor lattice structure and potential
correlation. Low period (A) is not directly measured; the test is distinct from uniform distribution tests (C) or
• The generator's period is less than 2^10. (Incorrect):
This option is not appropriate. A poor result means points lie on few widely spaced hyperplanes, indicating poor lattice
structure and potential correlation. Low period (A) is not directly measured; the test is distinct from uniform distribution
tests (C) or autocorrelation (D) though related
• The generator fails the chi-square goodness-of-fit test. (Incorrect):
This option is not appropriate. A poor result means points lie on few widely spaced hyperplanes, indicating poor lattice
structure and potential correlation. Low period (A) is not directly measured; the test is distinct from uniform distribution
tests (C) or autocorrelation (D) though related
• The sequence exhibits a long positive autocorrelation at lag 1. (Incorrect):
This option is not appropriate. A poor result means points lie on few widely spaced hyperplanes, indicating poor lattice
structure and potential correlation. Low period (A) is not directly measured; the test is distinct from uniform distribution
tests (C) or autocorrelation (D) though related
Page 3
, Question 3 of 107
In a simulation of a queueing system, you want to reduce the variance of the estimated mean waiting
time. Two candidate techniques are antithetic variates (AV) and control variates (CV). Assuming
perfect negative correlation for AV and a highly correlated control variate for CV, which statement is
correct regarding their efficiency improvement?
A. AV always yields greater variance reduction than CV because it requires no additional modeling.
B. CV can achieve arbitrarily high variance reduction if the control variate is perfectly correlated with the output.
C. AV reduces variance by using complementary random numbers, while CV uses known information about
related variables.
D. Both techniques are applicable only in terminating simulations, not steady-state.
The correct answer is:
Correct Action: AV reduces variance by using complementary random numbers, while CV uses known
information about related variables.
Rationales
• AV reduces variance by using complementary random numbers, while CV uses known information about
related variables. (Correct):
This is the correct action. Antithetic variates induce negative correlation between two runs using complementary random
numbers; control variates exploit correlation between the output and a known variable. Perfect correlation can make CV
extremely effective, but not arbitrarily (B is overstated). AV can be applied to steady-state too (D is false). A is false
• AV always yields greater variance reduction than CV because it requires no additional modeling. (Incorrect):
This option is not appropriate. Perfect correlation can make CV extremely effective, but not arbitrarily (B is overstated).
AV can be applied to steady-state too (D is false)
• CV can achieve arbitrarily high variance reduction if the control variate is perfectly correlated with the
output. (Incorrect):
This option is not appropriate. Perfect correlation can make CV extremely effective, but not arbitrarily (B is overstated).
AV can be applied to steady-state too (D is false)
• Both techniques are applicable only in terminating simulations, not steady-state. (Incorrect):
This option is not appropriate. Perfect correlation can make CV extremely effective, but not arbitrarily (B is overstated).
AV can be applied to steady-state too (D is false)
Page 4
Answers - 2026 Update | 100% Correct - GT. - 107 Questions and
Answers Already Graded A+ Premium Exam Tested And
Verified
Subject Area ISYE 6644-ASY, OAN Exam 1 Simulation | Questions and Answers - 2026
Update | 100% Correct - GT.
Description Comprehensive examination on ISYE 6644-ASY, OAN Exam 1 Simulation |
Questions and Answers - 2026 Update | 100% Correct - GT..
Expected Grade A+
Total Questions 107
Duration 3 hours
Learning Outcomes 1. Demonstrate mastery of core concepts
Accreditation Aligned with US university standards.
Page 1
,Question 1 of 107
In steady-state simulation output analysis, the batch means method is used to estimate the variance
of the sample mean. If the batch size is too small, the resulting confidence interval may be unreliable
because the batch means are correlated. Which condition is essential for the batch means method to
produce a valid confidence interval?
A. The number of batches must be at least 30 to ensure normality.
B. The batch means must be approximately independent and identically distributed.
C. The initial transient bias must be removed via a warm-up period.
D. The batch size must be equal to the number of replications.
The correct answer is:
Correct Action: The batch means must be approximately independent and identically distributed.
Rationales
• The batch means must be approximately independent and identically distributed. (Correct):
This is the correct action. The batch means method relies on the batch means being approximately i.i.d. If batch size is too
small, autocorrelation among batch means persists, invalidating the standard variance estimator and confidence interval.
Removing initial transient (C) is necessary but not the core condition for the batch means' validity.
• The number of batches must be at least 30 to ensure normality. (Incorrect):
This option is not appropriate. i.d
• The initial transient bias must be removed via a warm-up period. (Incorrect):
This option is not appropriate. i.d
• The batch size must be equal to the number of replications. (Incorrect):
This option is not appropriate. i.d
Page 2
,Question 2 of 107
A linear congruential generator (LCG) has modulus m = 2^31 - 1, multiplier a = 950706376, and
increment c = 0. The spectral test is applied to assess the generator's lattice structure in k
dimensions. Which of the following best interprets a poor spectral test result in 2 dimensions?
A. The generator's period is less than 2^10.
B. Points (u_i, u_{i+1}) fall on a small number of parallel hyperplanes.
C. The generator fails the chi-square goodness-of-fit test.
D. The sequence exhibits a long positive autocorrelation at lag 1.
The correct answer is:
Correct Action: Points (u_i, u_{i+1}) fall on a small number of parallel hyperplanes.
Rationales
• Points (u_i, u_{i+1}) fall on a small number of parallel hyperplanes. (Correct):
This is the correct action. The spectral test measures the maximum distance between parallel hyperplanes covering all
points. A poor result means points lie on few widely spaced hyperplanes, indicating poor lattice structure and potential
correlation. Low period (A) is not directly measured; the test is distinct from uniform distribution tests (C) or
• The generator's period is less than 2^10. (Incorrect):
This option is not appropriate. A poor result means points lie on few widely spaced hyperplanes, indicating poor lattice
structure and potential correlation. Low period (A) is not directly measured; the test is distinct from uniform distribution
tests (C) or autocorrelation (D) though related
• The generator fails the chi-square goodness-of-fit test. (Incorrect):
This option is not appropriate. A poor result means points lie on few widely spaced hyperplanes, indicating poor lattice
structure and potential correlation. Low period (A) is not directly measured; the test is distinct from uniform distribution
tests (C) or autocorrelation (D) though related
• The sequence exhibits a long positive autocorrelation at lag 1. (Incorrect):
This option is not appropriate. A poor result means points lie on few widely spaced hyperplanes, indicating poor lattice
structure and potential correlation. Low period (A) is not directly measured; the test is distinct from uniform distribution
tests (C) or autocorrelation (D) though related
Page 3
, Question 3 of 107
In a simulation of a queueing system, you want to reduce the variance of the estimated mean waiting
time. Two candidate techniques are antithetic variates (AV) and control variates (CV). Assuming
perfect negative correlation for AV and a highly correlated control variate for CV, which statement is
correct regarding their efficiency improvement?
A. AV always yields greater variance reduction than CV because it requires no additional modeling.
B. CV can achieve arbitrarily high variance reduction if the control variate is perfectly correlated with the output.
C. AV reduces variance by using complementary random numbers, while CV uses known information about
related variables.
D. Both techniques are applicable only in terminating simulations, not steady-state.
The correct answer is:
Correct Action: AV reduces variance by using complementary random numbers, while CV uses known
information about related variables.
Rationales
• AV reduces variance by using complementary random numbers, while CV uses known information about
related variables. (Correct):
This is the correct action. Antithetic variates induce negative correlation between two runs using complementary random
numbers; control variates exploit correlation between the output and a known variable. Perfect correlation can make CV
extremely effective, but not arbitrarily (B is overstated). AV can be applied to steady-state too (D is false). A is false
• AV always yields greater variance reduction than CV because it requires no additional modeling. (Incorrect):
This option is not appropriate. Perfect correlation can make CV extremely effective, but not arbitrarily (B is overstated).
AV can be applied to steady-state too (D is false)
• CV can achieve arbitrarily high variance reduction if the control variate is perfectly correlated with the
output. (Incorrect):
This option is not appropriate. Perfect correlation can make CV extremely effective, but not arbitrarily (B is overstated).
AV can be applied to steady-state too (D is false)
• Both techniques are applicable only in terminating simulations, not steady-state. (Incorrect):
This option is not appropriate. Perfect correlation can make CV extremely effective, but not arbitrarily (B is overstated).
AV can be applied to steady-state too (D is false)
Page 4
