DEPARTMENT OF STATISTICS AND DATA SCIENCE

STAT 6130 Fall 2022

Regression Analysis for Business
Syllabus

Professor Abraham (Adi) Wyner [email protected]

303 Academic Research Building
215 898-2439 (you can leave a voice message here)
@adiwyner twitter

Teaching Assistants:

Ryan Gross (PhD student) [email protected]. Office Hours: Fridays 12-1pm via Zoom
Kavya Garikipiti (2Y MBA) [email protected]. Office Hours: Tuesdays 6-7pm (zoom)
Ivanna Pearlstein (2Y MBA) [email protected]. Office Hours: Tuesdays 6-7pm (zoom,
same link as Kavya)

________________________
Source material

Required
• JMP 16 (Accessible for free through canvas)
• Pekoz, Erol Managers Guide to Statistics available at the bookstore
https://upenn.bncollege.com/my-account/rental-order/orders
• Class Notes (canvas)

Additional Reading Suggestions

• Sall, Creighton, Lehman, JMP Start Statistics, 5th Edition, SAS Institute.
• Freedman, Pisani and Purves, Statistics, 4th edition, Norton.
• Keller, Statistics for Management and Economics, 10th edition, 2014, South-
Western Cengage Learning.
• Stine and Foster, Statistics for Business, Addison Wesley.

6130 Syllabus - 1
Course Pedagogy

The traditional way of teaching a statistics class is lecture. The professor presents
formulas, justifies the formulas and then perhaps illustrates their use by solving problems
in class. Students ask questions about confusing points.

Although there will be plenty of lecture (some asynchronous), actual class time will also
feature in-class activities, student problem-solving, and discussions. Class will not meet
in person on September 26th and October 5th (Jewish holidays). I will post videos
which can be watched anytime.

For the course to work as well as possible, it is important that you prepare for class.
Before every class there is a reading and self-study problems (not submitted). It is highly
recommended that you prepare for class by do the reading first and then try to solve the
problems. Solutions to all the self-study problems are posted to Canvas. If you properly
prepare the lecture will be super clear and easy to follow. Towards the end of class there
will be breakout problems. This will be submitted but not graded (the TAs will check that
they have been submitted on time.) No credit will be given to late submissions.

The class is organized into modules on canvas where all the assignments posted.

Background

The statistics background for students in STAT 6130 is quite variable. Some of you have
never had any courses and others have had many. I will assume this is your first course in
statistics. For those of you who are more advanced, I will sprinkle in optional material in
“math supplement” designed to stretch you a bit.

The Wharton MBA program has a strong analytical focus, and the Statistics 6130 class
has always reflected that orientation. Sometimes the urgency to teach skills obscures the
need for conceptual understanding. To that end, I am assigning the book “Manager’s
Guide to Statistics” by Boston University Statistics Professor Erol Pekoz. The book is
highly conceptual; it is in many ways a managerial version of class text “Statistics” by
Freedman, Pisani and Purves (who were my teachers and advisors at Berkeley). I think
you will enjoy the book. If you are new to statistics (or it has been years and years) the
reading will be an essential part of the learning process.

JMP (pronounced “jump”) is the computer package we’ll use extensively for statistical
calculations and graphics. An essential component of 6130 will be homework that use
JMP.

Why use JMP?

There are many statistical packages, including SAS, Minitab and even Excel (if you are
generous). JMP has a growing user base, but it is not likely to be the tool of choice when
you return to the work force. But JMP has many advantages. It is extremely powerful.

6130 Syllabus - 2
Indeed, its most recent versions are equipped with an unsurpassed suite of artificial
intelligence tools including natural language processing, machine learning and internet
data acquisition. It can be used entirely through a “point and click” interface which is
super easy and highly conducive to exploration. Thus, you can concentrate on
understanding. You will also be doing very powerful analyses very quickly. It is also
used in many upper-level stat classes at Wharton and UPenn.

Can I learn R?

The book has optional material that teaches you to use the free statistical programming
language R. If majoring in Statistics is something you are seriously thinking about, then
learning to use R will be important. Use of R is optional in this course, but it will be
supported by the TAs. You may install and experiment with the free R software package
and the corresponding R studio editor using the two links here:

• R: https://cran.rstudio.com/
• R studio: https://www.rstudio.com/products/rstudio/download/

Class

Classes will begin with a lecture and often a problem or two sometimes based on the
reading. Then we will “flip” the class after about 50-60 mi and break your teams to
complete a worksheet of problems. We will then regroup and time permitting a team will
present their solution to the class. If time does not permit, then solutions will be posted or
solved quickly by the professor. You have been assigned seats according to your
learning teams.

Attendance and Videos

Use the attendance app to record your participation. If you miss class, you can submit
reasons and excuses. If you would like to watch the video of class, you need to ask the
TAs for access (which will be granted only for excused absences). There is no check-in
when there is no-class and not on the dates of the in-class exam. For information, see
(For Students) Wharton Attendance, Absence, and Video Requests Module

6130 Syllabus - 3
 Lecture Date Key Topics
1 How not to be fooled with data!
August 29 Causality, confounding and Controlled Studies. The method of comparison.
2 The Histogram
August 31 Portraying variation in the data. Summary measures of the center and spread
of the Data (median, mean and IQR). SD and Empirical Rule

3 More Data Visualization: Box plots, mosaic plots,

Sep 7 More Data Summary Examples

4 Normal Distribution:
Sep 12 A very useful tool for understanding data and making predictions.
QUIZ
5 Telling Statistical Stories with Graphs:
Sep 14 The graph-builder in JMP.

6 Correlation
Sep 19
7 More Correlation
Sept 21
QUIZ
8 Regression Equation, Regression Residuals
Sep 26 NO IN PERSON CLASS: WATCH VIDEO
Rosh Hashanah
9 Multiple Regression Continued
Sept 28 Fitting hyperplanes to data. Interpreting Regression Equations. Marginal
and Partial Slope.
10 Multiple Regression Continued
Oct 3 Dummy Variables, Interactions
11 Multiple Regression Examples. Transformations
Oct 5 NO IN PERSON CLASS: WATCH VIDEO
Yom Kippur
12 Exam I
October 10

6130 Syllabus - 4
 13 Probability: a “language” to reflect uncertainty
Oct 24 Conditional Probability and Bayes Rule
14 Random Variables and their Expected Value and SD
Oct 26
15 Sampling variation:
Oct 31 The law of averages and the standard error of the sample percentage.

16 Central Limit Theorem of Statistics

Nov 2
QUIZ
17 Confidence Intervals
Nov 7
18 Hypothesis Testing
Nov 9 One Sample tests, the p-value.
19 Hypothesis Testing:
Nov14 Two Sample tests: The “A/B” test

20 Confidence Intervals in Single Variable Regression

Nov 16 The p-value of a regression.
QUIZ
21 Confidence Intervals in Multiple Regression
Nov 28
22 Building a regression model
Nov 30 Stepwise regression, data mining.
23 How to diagnose problems in regression and what to do about them
Dec 05 Wrap-Up !!
Optional
Quiz 5
24 Exam II
Dec 07

Dec 21 Final Project Due

6130 Syllabus - 5
Homework, Quizzes and Exams
• There will be 6 homework assignments and one final exam-data-analysis project.
(5-10 hours of work). Brevity is encouraged! The final data analysis project is
individual. Assignments 3 and 5 are individual. Assignments 1,2,4, and 6 are
with your teams.
• There will be 4 short quizzes. These will be multiple choice. There will be a fifth
quiz that is a make-up for anyone who missed a quiz earlier in the semester.
• There will be two 1.5 hour exams in class.
• One week grade query maximum from the time work has been handed back (go to
Ryan Gross, TA).

Many problems will be from the book. All datasets required are posted to canvas. Some
of the homework assignments must be done individually and submitted online by the due
date and time. You may discuss the problems with other students in the class, but your
final write-up should be done individually or team and then uploaded to the
“Assignments” section of the course website. Group assignments are submitted once and
there is one grade for the entire group. The content of your homework is to be generated
by you (or your team) and you alone. If you copy work from another team or from
another source you not only will get severely penalized, you (and your team) will be
referred to the Office of Student Conduct for review.

Pre-Class Work: Self-Study and Reading

Before most classes, there is a reading assignment from the text and “self-study
problems” many based on the readings and lectures. These will not be graded, and they
will be simple and they do not need to be submitted.

Breakout Problems

After every lecture, there will be breakout problems assigned to be completed in groups
during class with students sometime presenting their solutions. You are required to
submit breakout problem solutions online. You can either type solutions directly into
canvas or upload a file or a photo. Completed submissions will get one point (regardless
of correctness). Submissions are due on the day they are assigned.

Teaching Assistants (TAs)

TAs for Stat 6130 will hold office hours throughout the course. Times and locations will be
posted on canvas. You can refer questions to any of your TAs. Questions about regrades
should be taken to your PhD student TA, Ryan Gross ([email protected]).

6130 Syllabus - 6
Class Participation Information

I will construe class participation broadly. There will be plenty of opportunity to

participate in class, you don't need to do them all equally well.

1. Answer questions and ask questions during lectures.

2. Submit breakout room solutions
3. Contribute to discussions on Canvas
4. Attend class consistently

Grading

1. The tests and the quizzes are in-class, taken on a laptop using the Lockdown
browser. They are closed book, closed notes.
2. Tests last one hour and thirty minutes
3. Quizzes last 10 minutes. There will be a “fifth” quiz that is optional and can be
used to replace the lowest quiz or used as a make-up, if an in-class quiz is missed.
4. Late homework is penalized by 25% for up to 24 hours late and 50% for up to 48
hours late. At that point solutions will be posted and no further submissions will
be accepted. Any homework grade queries must be made via email, to Ryan
Gross within one-week of the solutions being posted.
5. There are no extra credit opportunities in the course.

Grades for the course will be based on the following components

• Test 1: 20%
• Test 2: 25%
• Quizzes: 15%
• Homework: 20%
• Take-home final exam data-analysis project: 10%.
• Attendance and participation: 10%.

Makeup

There is no reason to miss a test or a quiz as they are scheduled in class, and students
should not schedule anything that conflicts with class time, especially on quiz and exam

6130 Syllabus - 7
dates. Please plan accordingly. Exceptions will only be made for illness, marriages, death
etc.

6130 Syllabus - 8