Cross Section Econometrics Syllabus [ECON 407]

Due to family health issues, I am unable to offer this class in person. This class is being taught as "Remote Asynchronous". This means, I will be recording lectures per lecture day where you can watch them on your schedule. Office hours (and a first day "live" zoom session) will occur over video conference at appointed times.

In this course we will explore econometric techniques for testing microeconomic theories at the individual or firm level. With the advent of computers and associated data on economic behavior, the past few decades has seen an explosion of applied economic research using a wide range of techniques for this type of data- termed cross section data and cross section econometrics. In this course, you will learn about these techniques, will learn to be an educated consumer of econometric research, and will apply these techniques to real data. We will also derive many of the properties of the statistical techniques used in this course, but primarily at the end of the course you will

• Understand the strengths and weaknesses of cross section techniques
• Know how to test the validity of modeling assumptions
• Know the proper econometric technique for a wide variety of Cross-Section settings

• Lectures : All lectures are pre-recorded and available on the course google drive share. If you have registered for the class you should be able to access the videos linked below. Note that these videos are for your own personal use in this class and all rights for the videos and other course materials are reserved by me (Copyright 2020 by Rob Hicks. All rights reserved.). Here is a list of notes and videos (in order) for each week of class:

  Unit	Week	Video Subject and Link	Running Time (minutes)
  Introduction	1 (Aug 19)	Recording of Zoom Kickoff	35.03
  		Introduction and Logistics1	25.12
  		Gitlab Demo	8.60
  		Markstat Installation Demo	10.50
  		Markstat Demo	15.02
  		Jupyter Demo2 (Optional)	7.07
  	2 (Aug 24)	Linear Algebra Review	47.27
  		Linear Algebra 2	2.63
  Ordinary Least Squares		Intro	17.48
  		Linear Algebra	23.68
  		Solving for \(\mathbf{b}\) (and this one)	54.78
  		Implementing \(\mathbf{b}\) in Stata	26.03
  	3 (Aug 31)	Assumptions	18.28
  		Unbiasedness	25.83
  		Variance of \(\mathbf{b}\) (and this one)	30.62
  		Unpacking variance/covariance matrices	14.25
  		Implementing \(VAR(\mathbf{b})\)	11.03
  		Replication in Stata	28.47
  		Heteroskedasticity	31.30
  	4 (Sep 7)	Implementing Heteroskedasticity	10.20
  		Testing for Heteroskedasticity	27.62
  		Heteroskedasticity in Stata	18.63
  		Clustered Standard Errors	7.18
  	(Sep 10)	Mid Term 1	N/A
  		Confidence Intervals	13.00
  Endogeneity	5 (Sep 14)	Introduction	22.68
  		Approaches for fixing problem	27.37
  		Properties of a good instrument	47.85
  		Estimating \(\beta\) with the Instrument	39.25
  		Testing for Relevancy and Endogeneity	27.15
  	6 (Sep 21)	Midterm 1 Discussion	19.07
  		IV Regression in Stata 1	28.00
  		IV Regression in Stata 2	35.00
  		IV Regression in Stata 3	29.43
  		More than 1 Instrument	53.72
  	7 (Sep 28)	IV Regression in Stata 4	30.62
  		Problem Set 2 Discussion	27.28
  Panel Data: Intro		Introduction	29.07
  		Pooled OLS, Random Effects	32.25
  		Introduction to Fixed Effects	37.70
  Panel Data: Random Effects	8 (Oct 4)	Random Effects Error Structure 1	45.20
  		Random Effects Error Structure 2	41.73
  Panel Data: Fixed Effects		Fixed Effects Dummy Variable Approach	46.57
  		Fixed Effects De-meaning estimator	31.32
  	9 (Oct 11)	Fixed Effects First Differences	41.00
  		Panel Applications in Stata	52.07
  		MidTerm 2 Overview	18.47
  	10 (Oct 18)	Panel Data and Difference in Difference	38.43
  Maximum Likelihood		Introduction to Maximum Likelihood	30.57
  		Limited Dependent Variables and OLS Fail	10.62
  		MLE Hands-On	37.48
  		Stata and MLE	36.42
  	11 (Oct 25)	MLE Properties	31.52
  		Nested Hypothesis Testing 1	42.45
  		Nested Hypothesis Testing 2	22.87
  		Non-Nested Hypothesis Testing	36.53
  Probit/Logit	12 (Nov 1)	Introduction	12.28
  		Building the Likelihood Function	42.53
  		Interpreting Parameters	33.73
  		Stata 1	30.78
  		Stata 2	36.68
  		Wrapup	10.70
  Censoring/Truncation	13 (Nov 8)	Introduction	30.50
  		Truncation as an Endogeneity Problem	21.03
  		Truncated Normal I	20.27
  		Truncated Normal II	22.92
  		Mills Ration as a Correction Factor	20.58
  		Truncated Regression	12.92
  		Tobit Model	42.22

• The list of videos will be extended as they are produced. Please note that the first few weeks of class have less than ideal (but still ok with headphones) audio/video quality and later videos are improving in this regard.
• Office Hours : Thursday 3:30 - 4:50pm or by appointment. I am here to help with questions you bring in zoom (logistics announced weekly via email). Office hour sessions won't be recorded and are not mandatory.

• Asking Questions: Substantive questions about course material or coding in Stata must be submitted to the Cross Section issue tracker at https://gitlab.com/robhicks/econ407. Before using the issue tracker, you must create an account on gitlab (free). Use your William and Mary email address so that I can add you to the user list. Part of Problem Set 1 requires you to file an issue at this site, so create your account ASAP once you receive an invite from me. Do not post proofs or code on the gitlab site that effectively answers an exam or problem set question. Use email for these types of questions.

• Email Policy : For other types of questions like setting up meetings, grade questions, logistical issues, etc., I will respond to emails but only if they contain the tag [ECON407] in the subject line or you use the email exactly as specified above. If not, the google will likely delete your email. Substantive questions about course material/coding should be posted to gitlab unless you feel that the code or information you provide in the issue tracker will answer the question.

• Grades: Your grade will be based on five exercises (1 @ 2% and 4 @ 7% each), mid-terms (2 @ 22% each), a final exam (26%).
  • The problem sets will consist of a mix of theoretical and practical econometrics and should be considered serious time-consuming undertakings. In each (after the first one), you will be given a dataset and will need to conduct an econometric analysis thinking critically about which technique to employ as well as key tests that should be run. Your document should include clear interpretations of your results, tables with clear variable names, and be well-formatted with code, tables, and writeup combined in a convenient (for me) way. You will be responsible for posting to the blackboard assignment a markstat file that completely generates your analysis: a flowing narrative containing code, writeup, and results. For any part of the problem sets requiring hand-written math excercises, you can turn those in with your pdf but the response must be clearly marked for the question and must be in proper order. Early on in class, I will demo how this works for stata. The handout for reproducible research can be found here and is produced by this stata source code in stata using markstat. Also, a markstat demo is provided for showing the workflow that I demonstrated in class. Note the special considerations for Mac users at the end of the video.

  • Grades will be awarded based on standard grading scales after a curve is applied (if necessary):

    Letter grade	Percentage of points after curve
    A	93–100%
    A−	90–92%
    B+	87–89%
    B	83–86%
    B−	80–82%
    C+	77–79%
    C	73–76%
    C-	70–72%
    D+	67–69%
    D	63–66%
    D-	60–62%
    F	0–59%

    For a typical class year, the points required for an "A" in the course before the curve is applied is approximately 84-85% of the total available points.

• The mid-terms are scheduled for September 10 and October 15. Unfortunately these times are fixed and can't be rescheduled. Furthermore, by university policy I am unable to change the final exam time. If these exam times don't work for you please drop the course.

• Final Exam: University policy will not allow me to reschedule the final exam (see the Dean of Students for exceptions).

• Course assignments must be turned in on time. Late work will be accepted for up to two additional days (with Saturday and Sunday counting as 1 day in total) with a letter grade deduction for each late day. After two days, late assignments will not be accepted. See below for some examples:

  Due Date	Turned in	Your Grade	Your Grade after Penalty
  Tuesday	Thursday	A	C
  Thursday	Saturday or Sunday	A	C
  Tuesday	Friday	A	F (not accepted)
  Thursday	Monday	A	F (not accepted)

No hardcopies of any assignments will be accepted this semester. Items slipped under my door or placed in my mailbox will not be retrieved or graded and will not count towards successful and timely completion of course assignments. All assignments must be submitted via the appropriate assignment in blackboard by the time stipulated in the blackboard assignment item.

I am happy to discuss questions you have about your grade on class assignments. Any questions you have regarding a potential grade change on an assignment must be cleared up within 1 week of receiving your work back from me. The only exception to this policy is if I made an arithmetic or data entry error in adding your score up and entering it into blackboard. I will not entertain grade questions at the beginning of or following a class. These need to be handled in my office.

All course materials are available on my website for this course at the links listed below. I will only be using blackboard for posting grades and problem set solutions.

I am no longer requiring any books for this class. However, there are two I highly recommend particularly if you plan on attending graduate school in Economics or Political Science. First is Greene's Econometric Analysis. Second is this one by Wooldridge which is recommended but not required. Both of these books are expensive and you can find older editions at less than half the price. For your convenience, beside each topic is the relevant parts of Greene and Wooldridge although these readings are not required.

The order of topics in the reading list below does not necessarily reflect the order topics will be covered in class.

We will make extensive use of Stata. You may want to buy your own copy of the software (through the Grad Plan at https://www.stata.com) and the cost is around $100. Alternatively, a College-owned version can be run remotely on a Unix machine (stat.wm.edu) from your laptop through X11 or other means. If you decide to go this route, know that the setup is time consuming and you will likely need assistance from IT. If you plan on using stat.wm.edu, please deal with these logistics during the first week of class. Additionally, there are computer labs around campus (e.g. Morton 240) where stata is accessible.

We will make some use of the main computer in the classroom for much of what we do together in class. The data for course examples and problem sets will be available on the web for the duration of the course, but you should have archival copies and working backups of all of the programs you have written to take the data from raw to final form. All computer work you do in this class must self-generate the full analysis and write-up using markstat as described on the first day of class.

Finally, I want to define acceptable collaboration. In this course, I want you to think for yourself in applying these techniques. Using resources outside of class and properly citing it, is perfectly with-in bounds. Taking .do or .stmd files from your fellow classmates or off of the web and treating them as your own work is not acceptable. Receiving assistance at every critical modeling step is also not acceptable. Asking a classmate about clarification of stata syntax (e.g.- I forgot how to ask for robust standard errors, could you help me with that?), is fine.

Be aware that all assignments submitted to blackboard are automatically scanned by Turnitin and are compared against the assignments for all current and past Cross Section classes as well against information from the web.