ISYE 6644 Summer 2024 Final Exam Graded A

If f(x, y) = cxy for all 0 < x < 1 and 1 < y < 2, where c is whatever value makes this
thing integrate to 1, then X and Y are independent random variables - ANSWER
TRUE. (Because f(x, y) = a(x)b(y) factors nicely, and there are no funny limits.)

In our Arena Call Center example, it was possible for entities to be left in the system
when it shut down at 7:00 p.m. (even though we stopped allowing customers to enter
the system at 6:00 p.m.). - ANSWER TRUE — because of the small chance that a
callback will occur.

An entity can be scheduled to visit the same resource twice, with different service
time distributions on the two visits! - ANSWER TRUE

Arena has a built-in Input Analyzer tool that allows for the fitting of certain
distributions to data. - ANSWER TRUE

The Kolmogorov-Smirnov test can be used both to see (i) if data seem to fit to a
particular hypothesized distribution and (ii) if the data are independent. - ANSWER
FALSE. (It's just a goodness-of-fit test.)

Welch's method is a graphical technique to estimate truncation (initialization bias)

points for steady-state simulation. - ANSWER TRUE

If you are using a ranking-and-selection procedure and two competitors happen to

fall within the indifference-zone, then you don't really care too much which one you
end up selecting. - ANSWER TRUE. That's why it's called the IZ!

TRUE or FALSE? Sequential ranking-and-selection procedures are designed to stop

early if one alternative seems to be way out in front of the others. - ANSWER
TRUE

Suppose, when designing a ranking-and-selection procedure, you have decided to

increase the desired probability of correct selection compared to a previous run of
the procedure. What can you expect? (a) Sample sizes that are about the same
(b) Larger sample sizes
(c) Somewhat lower achieved Pr(CS)
(d) A larger indifference zone
(e) Lower confidence in your selection - ANSWER b

In Arena, you can use a single DECIDE module to route customers to that one of
four servers having the smallest queue. - ANSWER TRUE
TRUE or FALSE? In Arena, you can use the expression EXPO(3) to generate
exponential interarrival times having a mean of 1/3. - ANSWER FALSE. The mean
of an EXPO(3) is 3.

TRUE or FALSE? In Arena, an alternative expression to generate exponential

interarrival times having a mean of 1/3 is -(1/3)*LN(UNIF(0,1)) - ANSWER TRUE.
This is the Inverse Transform Method to generate an observation coming from Exp(λ
= 3).

TRUE or FALSE? In Arena, it is possible to set the maximum capacity of a queue

from within a SEIZE module. - ANSWER FALSE. It is only possible to set a
maximum queue capacity from within a QUEUE block found in the Blocks template.
2

TRUE or FALSE? In Arena, the expression NORM(-1,2)+NORM(3,2) has the same

distribution as the expression NORM(2,4) - ANSWER FALSE. The above
expression evaluates to NORM(2,√ 8) since, in Arena, NORM(x,2) corresponds to a
normal random variate with a mean of x and a standard deviation of 2; so the
variance of this random variable is 22 = 4.

TRUE or FALSE? In Arena, it is only possible to use one resource at a time in a

PROCESS module. - ANSWER FALSE

TRUE or FALSE? In Arena, once an entity is defined, it is not possible to change the
priority in which it gets processed by a resource. - ANSWER FALSE

TRUE or FALSE? In Arena, it is not possible to use a single RELEASE module to

release multiple different resources - ANSWER FALSE. We observed this in the
ReleaseHospitalRoomLater demo in HW7

What is the distribution given by the Arena expression UNIF(1,2)+UNIF(1,2)? -

ANSWER Tria(2,3,4).

An Arena ASSIGN module can be used to assign values to attributes and variables -
ANSWER TRUE

TRUE or FALSE? In Arena, there are several ways to terminate a simulation, for
example, when a given simulation run length has been reached, or when a given
number of entities have been created, or when a specific stopping criterion as be
met. - ANSWER TRUE

Suppose X is a discrete random variable that can take on four possible values: −0.1,
0, 2, and 4. The associated probabilities are, respectively: 0.3, 0.1, 0.2, and 0.4. Find
an expression in Arena to generate one observation coming from this p.m.f. -
ANSWER DISC(0.3, −0.1, 0.4, 0, 0.6, 2, 1, 4)
DISC(F(x1), x1, F(x2), x2, . . .)

Suppose there are 3 people in the line called joe.queue and 5 people in the line
called tom.queue. What is the value of the following Arena expression?
(NQ(joe.queue) > NQ(tom.queue)) + (TNOW >= 0 - ANSWER The logical
expression (NQ(joe.queue) > NQ(tom.queue)) is false (because 3 ≤ 5) and therefore
0. The logical expression (TNOW >= 0) is always true (since time is always at least
0) and therefore 1. Thus, 0 + 1 = 1

TRUE or FALSE? The Arena Call Center model discussed in class used a
nonhomogeneous Poisson process arrival process (with rates varying throughout the
day). - ANSWER TRUE

Regarding Linear Congruential Generators (LCG's), select the statement that is

correct. - ANSWER Both full periodicity and cycle length properties should be
considered when choosing a good LCG.

We are testing a new medication. Our null hypothesis, H0, is the assumption that the
medication is safe. If we reject this null hypothesis but the medication is actually
safe, what type of error have we committed? - ANSWER Since we have incorrectly
rejectedthe null hypothesis of safety, we have committed a Type I error

TRUE or FALSE? In Arena, it is not possible use a single DECIDE module to

randomly route customers to one of three or more destinations. - ANSWER FALSE

TRUE or FALSE? In Arena, you can use the expression EXPO(3) to generate
exponential interarrival times having a mean of 3. - ANSWER TRUE

TRUE or FALSE? In Arena, an alternative expression to generate exponential

interarrival times having a mean of 3 is -(1/3)*LN(UNIF(0,1)) - ANSWER FALSE.
This is the Inverse Transform Method to generate an observation coming from Exp(λ
= 3), which has a mean of 1/3.

TRUE or FALSE? In Arena, it is possible to set the maximum capacity of a queue

from within a QUEUE block found in the Blocks template. - ANSWER TRUE. In
fact, it's pretty much the only way.

What is the variance of the random variable given in the Arena expression NORM(-
1,3)+NORM(3,4)? - ANSWER : In Arena, NORM(x, y) corresponds to a normal
random variate with a mean of x and a standard deviation of y. So the variance in
question is 3 2 + 42 = 25.

TRUE or FALSE? In Arena, it is possible to grab more than one resource at a time in
a PROCESS module - ANSWER TRUE

TRUE or FALSE? In Arena, it is possible to change the priority in which entities get
processed by a resource - ANSWER TRUE

TRUE or FALSE? In Arena, it is possible to use a single RELEASE module to

release multiple different resources - ANSWER TRUE. We observed this in the
ReleaseHospitalRoomLater demo in HW7

What is the distribution given by the Arena expression UNIF(0,2)+UNIF(0,2)? -

ANSWER Tria(0,2,4)
TRUE or FALSE? An Arena ASSIGN module can be used to assign values to
attributes but not variables. - ANSWER FALSE. It can be used for both

Suppose X is a discrete random variable that can take on four possible values: −2, 1,
4, and 7. The associated probabilities are, respectively: 0.1, 0.3, 0.2, and 0.4. Find
an expression in Arena to generate one observation coming from this p.m.f. -
ANSWER DISC(−2, 0.1, 1, 0.4, 4, 0.6, 7, 1)

Suppose there are 3 people in the line called joe.queue and 2 people in the line
called tom.queue. What is the value of the following Arena expression?
NQ(tom.queue) ∗ NQ(joe.queue) + (TNOW < −1) - ANSWER The logical
expression (TNOW < -1) is false (since time is always at least 0). Thus,
NQ(tom.queue) ∗ NQ(joe.queue) − (TNOW < −1) = 6 + 0 = 6.

TRUE or FALSE? The Arena Call Center model discussed in class restricted all
servers to one specialty product each. - ANSWER FALSE. The use of resource
sets allowed certain servers to work on multiple products.

The runs "up and down" test is most often used to test for goodness-of-fit of
observations. - ANSWER FALSE. It's a test for independence.

Draw an Arena DECIDE block - ANSWER It's a sideways diamond.

In what Arena template would you find the Sequence spreadsheet? - ANSWER
Advanced Transfer.

In Arena, you can LEAVE a station. What is the analogous block to use to get into a
station? - ANSWER ENTER

Which method is not a viable way for an entity to LEAVE a resource? (i) Route; (ii)
Connect; (iii) Move; (iv) Transport; (v) Convey. (There may be more than one correct
answer.) - ANSWER Move

TRUE or FALSE? Suppose that U1, U2, . . . are truly i.i.d. U(0,1) random variables.
Then a 95% chi-square goodness-of-fit test for uniformity will incorrectly reject
uniformity of the observations about 5% of the time. - ANSWER TRUE. (That's
what Type I error is.)

TRUE or FALSE? Suppose Z1, . . . , Z6 are i.i.d. standard normal random variables
obtained by the Box-Muller method. Then P6 i=1 Z 2 i ∼ Erlang3(1/2). - ANSWER
TRUE. The sum of 6 i.i.d. Z 2 i random variables is a χ 2 (6), which is itself an
Erlang3(1/2).

TRUE or FALSE? If we want to generate X ∼ Pois(4.5), then it's better to use a

normal approximation than acceptance-rejection. - ANSWER FALSE. (The normal
approximation requires larger λ for the asymptotics to work and the efficiency to kick
in.)
TRUE or FALSE? The acceptance-rejection's majorizing function t(x) is usually a
p.d.f. - ANSWER FALSE. (It integrates to something > 1.)

If X1, . . . , Xn are i.i.d. Exp(1/9), what is the expected value of the sample variance S
2 ? - ANSWER E[S^2 ] = σ^2 = 81.

Find x such that e 2x = 1/x. (Get within two decimals) - ANSWER Bisection search
quickly reveals that x is about 0.4265

TRUE or FALSE? The square root of the sample variance is unbiased for the
standard deviation - ANSWER FALSE. (E[S 2 ] = σ 2 ; E[S] = σ.)

Which is usually a better way to deal with initialization bias in steady-state simulation
analysis: (i) make an extremely long run to overwhelm the bias, or (ii) perform
truncation? - ANSWER Truncate

If W(t) is a standard Brownian motion process and a < b, find Pr(W(a) < W(b)). -
ANSWER W(a) − W(b) is normally distributed with mean 0. This immediately yields
a probability of 1/2.

Suppose I use the method of overlapping batch means with sample size n = 10000
and batch size m = 500. Approximately how many degrees of freedom will the
resulting variance estimator have? - ANSWER Denote b = n/m = 20. You get
approximately 3 2 (b − 1) = 28.5 d.f. (Will also accept 3b/2 = 30 or anything
reasonably close.)

Suppose that I'm interested in selecting the most popular television show during a
particular time period. What kind of selection problem is this — (a) normal, (b)
multinomial, or (c) Bernoulli? - ANSWER b