DATA 420 — Modeling and Simulation

Professor: Stephen Davies

Spring 2024

Lecture:
TR 2–3:15pm, Farmer 026

Office hours (Farmer 044):

Mondays 12–2pm
Tuesdays 3:30–4:30pm
Wednesdays 10–11am
Thursdays 1–2pm

Final exam: Thu May 2nd, 3:30-6pm

http://stephendavies.org/data420

Computers are tools that can help us understand our world.

It’s true, of course, that computers do many other things as well: they store and
retrieve information, carry out financial transactions, and display 3-D graphics of
spaceships. For this reason, we’re sometimes likely to see computer programs princi-
pally in terms of what we can build with them, rather than what we can learn from
them.
This class, by contrast, is about seeking understanding, not about building applica-
tions or devices.
For a long time, there have essentially been two ways to gain knowledge. The first
one — theory — goes back to before Aristotle. We analyze our world, using our
reason, and come up with simplifications, abstractions, and categorizations that help
us make sense of it. Around the 17th century, a second method came on the scene:
experimentation. Instead of just speculating about how things behave, we subjected
Nature (and People) to principled testing, seeing if we could reliably and repeatably
predict how they will respond.
The 20th -21st centuries introduced a third way: modeling and simulation. This new
paradigm, made possible by the digital computer, helps us answer some questions

1
that were unanswerable previously. Estimating the energy in a nuclear chain reac-
tion? Plotting the trajectory of a hurricane? Predicting the outcome of a political
revolution? Anticipating how large a city can get before its crime rate increases?
None of these problems or thousands of others are solvable without adding this pow-
erful third weapon to our arsenal.
At the center of this new approach are abstract computational models that extrap-
olate outcomes from innumerable tiny elements and rules. And the complexity that
we see arising from lots of simple things interacting in seemingly simple ways is
breathtaking.
We technologists have a powerful skill set which can further our knowledge of society
and of the universe. The amazing thing is that without even getting our hands
dirty in a petri dish, the programs we write shed real light on real problems. Our
models and their simulations become the science in a very real way, as we leverage
the computer to attack the grand challenges of our age.

Course Objectives
• To give you a high-level overview of the techniques, promises, and limits of
computational modeling as a way to understand complex phenomena.

• To tie together the two halves of Data Science. Whereas a course like 419
(Data Mining) is about analyzing data to infer properties about the process
that generated it, 420 is about creating models of the process itself and seeing
whether the data it generates matches up. It’s kind of like 419 in reverse.

• To introduce you to some of the important modeling and simulation paradigms,

with a special focus on two: System Dynamics, and Agent-Based Modeling.
Also, to give you plenty of practice implementing and analyzing such models!

• Finally (and seriously) to improve your understanding of life in general and

make you a better thinker.

Student Learning Outcomes

After completing this course, students will be able to...

2
 • ...identify the different subfields within the modeling and simulation field, and
to explain their techniques, promises, and limits as a way to understand com-
plex phenomena.
• ...explain how modeling and simulation (DATA 420) – producing and analyzing
the data from a simulated model – is the inverse of data mining (DATA 419)
– reasoning from observed data backwards to a plausible model.
• ...design and implement programmatic models in both the Systems Dynamics
and Agent-Based Modeling paradigms, and accurately interpret their results.
• ...wield a new set of quantitative tools to assess uncertainty, think more criti-
cally, and form conclusions more reliably.

Rules of the game

1. There are absolutely, positively, NO stupid questions!! Your job is not to
already know everything before you start the course. Your job is to try hard
to learn, and part of that involves asking questions. I’m a nice guy, and I will
not ever belittle you, snub you, or make fun of you; and if anyone else does so
I will personally break both of their arms.
2. This class will be interactive. When I point at you in class, say your first name,
and be prepared to try and answer questions. (Don’t worry if you don’t know
all the answers.)
3. This class involves a good bit of reading. This is because the class involves a
good bit of learning, and with apologies to this video-centric generation, most
deep learning still comes through reading. Part of what you should get out
of this class is an improvement in your reading habits. I’ll work hard on my
end to try and make that happen, since I think becoming a more efficient and
effective reader is actually the most important thing you can get out of college,
period. What you need to do is have a good attitude about it and dedicate
yourself to rolling up your sleeves and digging in.
4. Don’t skip class. Just don’t. It’s bad form. I work hard to prepare for class,
to make it compelling and relevant. It hurts my feelings when you don’t come.
Plus you miss out on important stuff, and you’ll end up falling behind if you
skip lecture. So come every time. Come happy, fresh, excited, ready to think
and to participate.
5. Absolutely no laptops, cell phones, or other devices during class.∗
∗
Obviously “no laptops” doesn’t apply to any Zoom sessions we might have. The “no cell phones
or other devices” rule does, however.

3
 I’ve had students claim that they take notes on their laptop during lecture,
but even if it’s true, those things are way too big a distraction to you and your
fellow students to make it worth it. Just stay tuned in, because I move fast.
6. For any Zoom class that may take place this semester, you must have your
webcam on during the entirety of the lecture. If you don’t have a
working Webcam, buy one immediately.

The Honor Code and this course

For this course, all the work that you turn in for a grade must be solely
your own work, period. Specifically, this means:

• The reading checks must be taken without looking at the book and without
help from anyone.
• The Canvas quizzes must be taken alone, in a quiet place, without any form of
contact with anyone.
• You must write all your own Python simulations in their entirety.
I don’t mind if you chat informally about the assignments with your fellow
students, but you must not show anyone else your code nor look at
anyone else’s code. This includes people who are not in the class nor even
at UMW. I am happy to help you over email or in office hours about
whatever questions and problems you have.

Books
• Silver, N. (2015). The Signal and the Noise: Why So Many Predictions Fail –
But Some Don’t, New York, NY: Penguin Books, 2015.

One of true geniuses thus far in the 21st century, in my opinion, is the hugely in-
fluential Nate Silver of fivethirtyeight.com fame. He got his start forecasting
statistical outcomes for major league baseball players, of all things, but has since
done incredibly broad and ground-breaking work analyzing statistical models of all
kinds. His book The Signal and the Noise is hands-down one of the best books I
have ever read, and has literally changed the way I look at life. It is particularly

4
relevant to this class, as Silver has years of sober-minded expertise to share about
simulations and their predictions.
Silver has also been parodied on Saturday Night Live, so there’s that.

Late policy
No late work will be accepted this semester. Get your stuff in on time!

Grading
Grading this semester will be based on “experience points” (XP). As you complete
activities, you will earn XP towards your final total. XP can never be lost, only
gained, but you have to earn what you get (i.e., you don’t “start off with a 100%”
and lose points based on mistakes you make).
There will be opportunities to earn XP throughout the course. Some of these will
be spontaneous as the mood strikes me. Some you can earn by completing in-class
activities. Some may be in response to impressive things I see you do as the semester
progresses. The following opportunities, however, are guaranteed to be available to
you:

Guaranteed XP opportunities:
Activity Possible XP
One-minute reading checks 4 each
Eight open-book, open-note, timed Canvas quizzes 30 each
Seven programming assignments 40 each
Final exam (comprehensive) 100
Various and sundry others varies

The one-minute “reading checks,” by the way, are closed-book, open-note, and will
take place at exactly 2:00pm at the start of most class periods. These are intended
just to hold you accountable to the reading for the day, and should be easy as long
as you have done so.

5
Grading levels
Here are the levels you may achieve, together with the grade awarded (if any) and
the points necessary to reach!

Level Total XP Semester grade

The Doctor 700 A+
Mr. Spock 660 A
Dr. Hari Seldon 600 A–
Dr. Susan Calvin 560 B+
Shuri 520 B
Dr. Liet-Kynes 480 B–
Dr. Bruce Banner 440 C+
Dr. Ellie Satler 400 C
Dr. Jane Foster 360 C–
Dr. Miles Bennett Dyson 330 D+
Dr. Eleanor Arroway 300 D
Dr. Mark Watney 265
Dr. Stephen Falken 230
Felonius Gru 200
Dr. Emmitt “Doc” Brown 175
Buddy Pine 150
Dr. Joseph Albert Nefario 125
Dr. Sheldon Lee Cooper 100
Dr. Leonard Leakey Hofstadter 80
Dr. Peter Venkman 60
Dr. Bunsen Honeydew 40
Jimmy Neutron, Boy Genius 20
Beaker 0

6
Submitting programs
Rules for submitting programs will be given when the program is assigned. Most
of the time, you’ll be emailing me your program code as an attachment, and using
a specific subject line to distinguish it from my hordes of other email. Meeting the
deadline is a matter of sending your email before time expires.
Also, most of my homeworks are due at “midnight.” Here’s what “midnight” means:
if a homework is due “at midnight on Thursday,” then it is due after all of Thursday
has elapsed, and the clock strikes twelve. (In other words, this is good news: you
have all Thursday to work on it.)

Basis for determining mid-semester reports

For midterm progress reports, I look mostly at (a) whether you’ve been turning
assignments in (and preferably on time), and (b) quiz scores. If either or both of
these categories are lacking, it’s a sign of danger, and I will give you a “U” for your
mid-semester grade. Please don’t hesitate at all to come talk to me about this so we
can figure out how you can do better in the course.

Guidelines for class participation

I believe that students learn best when they participate wholeheartedly in all aspects
of the learning process. Hence while your grade will not be partially determined by
any “class participation score” per se, it is very much to your advantage, and very
much recommended, that you join in during class discussions, ask questions, and
make comments.

Disabilities
If you have a documented disability, please present me your letter from the Office of
Disability Resources and I’ll be happy to accommodate you.

7
Title IX Statement
UMW faculty are committed to supporting students and upholding the University’s Policy
on Sexual and Gender Based Harassment and Other Forms of Interpersonal Violence.
Under Title IX and this Policy, discrimination based upon sex or gender is prohibited. If you
experience an incident of sex or gender based discrimination, we encourage you to report
it. While you may talk to me, understand that as a “Responsible Employee” of
the University, I must report to UMW’s Title IX Coordinator what you share.
If you wish to speak to someone confidentially, please contact the confidential resources
below. They can connect you with support services and help you explore your options.
You may also seek assistance from UMW’s Title IX Coordinator; their contact information
can be found below. Please visit http://diversity.umw.edu/title-ix/ to view UMW’s
Policy on Sexual and Gender Based Harassment and Other Forms of Interpersonal Violence
and to find further information on support and resources.

Resources Confidential Resources

Ruth Davison, Ph.D. On-Campus
Title IX Coordinator Talley Center for Counseling Services
Lee Hall, Room 401 Lee Hall 106, 540-654-1053
540-654-5656
[email protected] Student Health Center
Lee Hall 112, 540-654-1040

Off-Campus
Empowerhouse (24-hr hotline)
540-373-9373

RCASA (24-hr hotline)

540-371-1666

Recording Policy
Classroom activities in this course may be recorded by students enrolled in the course
for the personal, educational use of that student or for all students presently enrolled in
the class only, and may not be further copied, distributed, published or otherwise used
for any other purpose without the express written consent of the course instructor. All
students are advised that classroom activities may be taped by students for this purpose.
Distribution or sale of class recordings is prohibited without the written permission of
the instructor and other students who are recorded. Distribution without permission is
a violation of copyright law. This policy is consistent with UMW’s Policy on Recording
Class and Distribution of Course Materials.

8
Accessibility statement
The Office of Disability Resources has been designated by the university as the
primary office to guide, counsel, and assist students with disabilities. If you receive
services through the Office of Disability Resources and require accommodations for
this class, please provide me a copy of your accommodation letter via email or during
a meeting. I encourage you to follow-up with me about your accommodations and
needs within this class. I will hold any information you share with me in the strictest
confidence unless you give me permission to do otherwise.
If you have not made contact with the Office of Disability Resources and have reason-
able accommodation needs, their office is located in Seacobeck 005, phone number
is (540) 654-1266 and email is [email protected]. The office will require appropriate
documentation of disability.

Basic needs security

Learning effectively and engaging wholly in class is dependent upon our basic security
and having our fundamental needs met: having a safe place to sleep at night, regular
access to nutritious food, and some assurance of safety. If you have difficulty affording
groceries or accessing sufficient food to eat every day, or if you lack a safe and
stable place to live, please contact Chris Porter, Assistant Dean of Students, at
[email protected]. Additionally, the Gwen Hale Resource Center is a free resource
on campus, providing food, toiletries and clothing to any member of our community.
It is open Monday, Tuesday and Friday from 1pm-6pm, on the 5th floor (floor A for
Attic) of Lee Hall, or [email protected]. Finally, you are always welcome to talk
with me about needs, if you are comfortable doing so. This will enable me to provide
any resources I may possess.

How to reach me
Come to office hours, see me after class, or e-mail me ([email protected]).

9
Calendar
(Note: the official calendar for the course, complete with assignment due dates, tests, etc., will be
maintained on the course website at http://stephendavies.org/data420.)

Week Topics Readings Due

1 Modeling and Simulation overview Silver: Preface & Intro

Modeling and Simulation principles Asimov: “Foundation”
2
The Python ecosystem Numerical calc. tutorial
System Dynamics modeling:
3 • Stock & flow diagrams Silver: ch. 1 HW #0
• Numerical calculus
4 Exponential and logistic growth patterns Silver: ch. 2
5 SD system patterns and analysis Silver: ch. 3 HW #1
SD models:
6 • Bass model of diffusion Silver: ch. 7
• SEIR
SD models:
7 • Lotka-Volterra Silver: ch. 4 HW #2
• Competitive exclusion
Path dependence
8 Silver: ch. 5
Stochastic systems
Probability in Python
9 Silver: ch. 8 HW #3
Random walks
10 Polya vs. Markov processes Prob/stat tutorial
Analyzing stochastic simulations
11 Silver: ch. 10 HW #4
Exploiting parallelism
12 Agent-Based Modeling concepts Silver: ch. 11
ABM models:
• Boltzmann
13 Silver: ch. 12 HW #5
• Granovetter
• El Farol
14 Grid-based and graph-based ABMs Silver: ch. 13
ABM models:
• Schelling
15 Silver: conclusion HW #6
• Local majority
• Binary Voter Model

10