BUEC333 Statistical Analysis of Economic Data (Fall 2017)

Fall 2014, Fall 2015, Fall 2016

My BUEC333 Policies

Syllabus
Textbook for the course: Stock and Watson (SW) 3rd edition updated. It has to be the updated edition .
- Problem sets will be assigned from the updated 3rd edition. These exercises are different from those in the 3rd edition.
- Solving the problems from the 3rd edition won't give you any credit.
If you need extra practice , check out the Shaum outlines. Google the book and you'll find free pdf's flying around.
- Shaum's outline on statistics contains a lot of solved problems, so if you are struggling with this class, consider solving as many problems as you can from both Stock and Watson and Shaum's.

I have a no email policy. This means that I will not be answering your emails.
- You can contact me by
  - (1) coming to my office hours -- TBD in WMC 2672
  - (2) seeing me immediately after lecture, or
  - (3) seeing me during the break that we always take during our 3-hour lectures.
Missing exams : There are no make-up exams. If you miss any exam, you will be assigned a grade of zero.
Asking for a regrade : If you would like to have your exam regraded, please, read and fill out this form.
- You have to fill out the form on the day you get your exam back, and submit both the form and the exam back to your TA.
- If you do not ask for a regrade the day you receive your exam back, your exam will not be considered for regrading. That is, you cannot ask for a regrade after you took the exam away from the classroom.
Submitting your work :
- Submit the homework assigned from the book individually, even if you work on it with a group. Add the names of the people in the group, but have everyone submit their own problem set. Submit your homework to your TA either right before the tutorial or at the end of the tutorial. The TA is there to answer any questions you may have about the current homework, but she is not required to solve your problem.
- You can submit the R assignments as a group.
- Add the number of your tutorial session on everything that you submit (homework, R assignments, or exams). If you do not do this, your work will not be graded, earning you a zero. No exceptions.

Software: R

R can be downloaded for free from its official website.
RStudio (interface for R) can be downloaded here
To learn R in R, check out swirl.
Entertaining intro to R here. Less entertaining intro to R here.
R by example.
A really great R resource.
Free book via SFU library. Check out the Intro, Basics, and Linear Regression.

Information on R labs:

R labs are held:
- Thursday 9:00-11:30
- Friday 8:30-10:30 and 15:30-17:00.
- in WMC 2506
Feel free to work in groups of max 3 people.
If you are struggling with the assignment, you are expected to go to the lab and work with the lab TA on the assignment.
When you submit your assignment, please, do the following:

Lectures mostly based on Stock and Watson (SW)

Part I: Probability Theory and Statistics

Week 1, 09/07:
- Lecture 1 and Ch 1 in SW
- Lecture 2 based on 2.1, 2.2, 2.3 in SW
Week 2, 09/14:
- Introduction to R. Bring your laptops, please.
Week 3, 09/21:
- Summary of the idea behind inference
- Lecture 3 based on 2.3, 2.5, 2.6, 3.1 in SW
Week 4, 09/28: Exam 1 on Part I
- covers Ch 1, 2 (without Central Limit Theorem), and Section 3.1
- formula sheet

Part II: Linear Regression

Past Exams

Past exams

Correlation versus causation

Guess the correlation game. [article]
Some funny stuff:
- spurious correlations
- pirates and global warming
- How the media interprets research papers here.
- Partying with the outliers
An article in WSJ: Do unemployment benefits cause unemployment?

Uncertainty in Economic Statistics

Article on how to communicate uncertainty of estimates in economics, Significance Magazine...
... and the article on which the article above is based

submit one assignment per group
write down the names of all people who have worked on the assignment. Each individual in the group will receive the same grade for the lab.
submit your R code together with your answers.

Week 5, 10/05:
- Lecture 4 based on Ch 3 in SW
- animation on the CLT
Week 6, 10/12:
- Lecture 5, based on Ch 4 in SW (no typed notes)
  - Non technical intro to regression [link]
  - Technical details for OLS estimators [link]
  - Interpretation of OLS parameters [link]
  - R-squared

If assignment of group/person A appears to be identical to that of group/person B, but A's name is not on B's assignment (or/and) vice-versa, then both A and B will receive an automatic zero (F) on the assignment.
Submitting identical assignments (even if answers to just one question are identical) but not giving credit is considered cheating. This will result in a grade of zero (F) for those particular assignments.
All students with identical, word-by-word assignments/answers will receive a zero.

Week 7, 10/19:
- Lecture 6 is a review for Exam 2
  - Law of Large numbers, Central Limit Theorem, Hypothesis testing for the mean, OLS with one covariate: estimation and inference
Week 8, 10/26 Exam 2 on Part II
- covers the end of Ch 3 (CLT) and Ch 4. See past exams.

Part III: Nonlinear Regression, Instrumental Variables

Week 9, 11/02:
- Lecture 7 based on Ch 6 (6.1 - 6.7) and Ch 5 (5.1 - 5.4) (no typed notes)
- Lecture 8 based on 8.1, 8.2 (dummy variables, nonlinear regression)
Week 10, 11/09:
- Lecture 9 based on based 6.1 (omitted variables) and on 12.1 and 12.5 (instrumental variables)
- Make sure you know Key Concepts 12.2 and 12.3.
- Youtube videos:
  - Intuition behind IV
  - Return to education example
  - IV and 2SLS
Week 11, 11/16:
- Review for exam
Week 12, 11/23: Exam 3 on Part III
- multivariate regression (the F-test), nonlinear regression (in particular log-log), dummy variables, and instrumental variables.
- Interpretation of slope parameter in OLS
Week 13, 11/30 (last day of classes)
- Lecture 10 on panel data and fixed effects