ARE 106: Econometric Theory and Applications

Summer Session II 2019

Instructor: Aleksandr Michuda (

Office Hours: Tuesday 2-4 or by appt (SSH TA Room)

TAs: Bret Stevens (

Discussion: Thurs 2:10-3:50 and 4:10-5:50 (Olson 141)

Office Hours: TBD

Class Website: Visit the canvas site for this course regularly for handouts and announcements, as well supplementary resources.

UCD Student resources: For answers to questions about academic support, health and wellness, career and internships, or campus community, go to


Course Objectives: You have all studied economic theory and statistics. This course brings together economic theory, mathematical tools, and statistics in order to do three things:

(1) Estimate economic relationships and predict economic outcomes, e.g., how does a change in price affect consumption? Can we predict your future earnings based on the education you are getting now?

(2) Generalize from a sample of data to whole populations, and

(3) Test hypotheses about cause and effect.

Additionally, our objective here will also be to understand how to accomplish these goals using computer programming. This will involve becoming more familiar with the command line on a computer and the basic ins and outs about correct way to code. We will be using Python for this course, which is a valuable skill to have in and of itself.

Please refer to the installation guide here to install the required software:

Prerequisites: ARE 100A, STA 103.

Lectures: Monday, Tuesday, Wednesday 4:10-5:50 in Hoagland Hall 168.


Computing: We will use Python for the computing in the class. This will involve installing and running the Anaconda Python Distribution. Homework will be done using Jupyter Notebooks, which is a useful tool for visualizing code and running estimations.

We will have several classes dedicated to learning how to code, so bringing a laptop on those days is recommended.

Homework and exams may also include questions on programming concepts as well as programming logic.


Reading: The textbook is Essentials of Applied Econometrics, by Aaron Smith and J. Edward Taylor. Oakland: University of California Press. Available on-line through UC Press:

eBook Access: Rather than buying the physical copy, you also have the option of buying the eBook version of the textbook. Check the canvas website for access.


Statistics Review: I assume that you remember and understand the statistics you learned in Stat 103 (or equivalent). Now is a good time to review that material. There are also reviews of stats for econometrics available on-line. A couple of examples are:


Assessment: There will be five homework assignments, a midterm, and a final exam. The midterm is scheduled for Monday, August 26th from 4:10-5:50 in Hoagland Hall 168 (i.e. during our regular class time) and the final will be on the last class that we meet, September 11th (during regular class time as well). This exam date is not negotiable. No consultation with any people, books, notes or other resources will be permitted during the exams. Any violations of this, or any other policy stated in this syllabus, will be dealt with according to University policy.

There are a total of 100 points available in this class. Your letter grade will be determined from your score out of 100.

Homeworks (50 points Total)

I will give you five homework assignments throughout the course. To get full credit on a question, you have to put in a good faith attempt on the question, or no credit will be given. These will include questions that will be pertinent for what we are doing and will give you practice for the exams.

The credit for the first homework, you will receive automatically, and it will be to set up Anaconda, Python and Jupyter notebooks, as well as any package dependencies that you might need so that you can do your analysis and homework through Jupyter.

Homework assignments will be submitted in Jupyter notebook format and will not be late. Since you have a full week to come to us and figure out how to make Jupyter work on your computer, failure to submit the homework on time and in Jupyter format will result in a missed homework.

The first homework assignment will be worth 10 points on your grade. The other four homeworks will be worth 13.5 points each. You can only receive 50 points through the submission of homework assignments, which presents an opportunity for you. If you do well enough on the three homework assignments, you might not need to do the last homework. 

Midterm and Final (25 points each -> 50 points Total)

The midterm will be worth a maximum of 25 points and the final exam will be worth 25. I will allow you to use your score on the final exam in place of your midterm (if this improves your overall score). This policy allows you to make up for a bad midterm by doing well on the final.

On the pages that follow, there is a course calendar. It has an approximate lecture schedule as well as exam dates and due dates for homeworks. It should be a useful resource for you. Please contact me if you have any questions, complaints or compliments about the class.

Aleksandr Michuda

August 2019




AUG 5 6 7 8 9

Lecture 1

Introduction and Stats Review


Reading: Ch. 1

Lecture 2

Simple Regression



Reading: Ch. 2

Lecture 3

Introduction to Coding and Python



AUG 12 13 14 15 16

Lecture 4

Multiple Regression

Reading: Ch. 3

HW 1 Due

Lecture 5

Multiple Regression

Reading: Ch. 3

Lecture 6

Python and Data



AUG 19 20 21 22 23

Lecture 7

Generalizing from a Sample

Reading: Ch. 4

HW 2 Due

Lecture 8

Properties of Estimators


Reading: Ch. 5

Lecture 9

Properties of Estimators


Reading: Ch. 5



AUG 26 27 28 29 30
Midterm Exam

Lecture 10

Hypothesis Tests and Confidence Intervals

Reading: Ch. 6

HW 3 Due

Lecture 11

Predicting in a Nonlinear World

Reading: Ch. 7



SEP 2 3 4 5 6
Labor Day (No Class)

Lecture 12


Reading: Ch. 8

HW 4 Due

Lecture 13

Sample Selection Bias

Reading: Ch. 10



SEP 9 10 11 12 13

Lecture 14

Identifying Causation

Reading: Ch. 11

HW 5 Due

Lecture 15

Instrumental Variables


Reading: Ch. 12

Final Exam