Empirical Analysis I (4 credits)PA 5031
Lecture 8: 11:15 ~ 12:30 Tuesday & Thursday HHH 25
Lab 10: 2:30 ~ 3:45TuesdayHHH 85
Lab 9: 1:00 ~ 2:15 Thursday HHH 85
Instructor & Teaching Assistants
Jason Cao, 295G Humphrey School, , 612-625-5671.
Office hour: 4-5 pm Tuesday and by appointment
Besides office hour, the best way to reach me is by email. Start with PA5031 and several-word summary of your questions in the subject. I may copy to all students if the questions are common.
Shunhua Bai (): 4-5 pm Monday
Alicia Valenti(): 4-5 pm Wednesday
Course Objectives
The objectives of this course areto help you
- evaluate empirical evidence in the media and scientific articles;
- comprehendprinciples of probability theory and statistics, apply them in research or policy analysis, and infer policy implications from statistical analysis;
- understand basics of sampling and survey administration; and
- establish a foundation for advanced statistics and survey design.
In this section, we will use examples in urban and regional planning, as well as public policy. We will use Stata to analyze data on land use, transportation, and health in the lab.
Textbook
Freedman, David, Robert Pisani, and Roger Purves (2007). Statistics, 4th edition. New York: Norton. ISBN 0-393-92972-8. (Most lectures are adapted from this book. So are most homework and exams.)
Utts, Jessica (2005). Seeing through statistics, 3rd edition. Belmont, CA: Thomson Brooks/Cole. (We will cover a few chapters of this book. It is not required to buy.)
Both books will be on reserve in the Wilson Library. Other readings will be posted on the course web.
Teaching Styles
Research shows that students learn more and remember what they learn much longer when they are active participants in the learning process. Be ready to participate in group discussions, think-pair-share, in-class exercises, and so on. The goal of these strategies is to facilitate your learning through engagement.
Questions in Class
I strongly encourage you to ask questions. Framing questions is part of the learning process. Some questions I will answer right away, because it is important to clear up a confusing point that is critical to our topic. Some questions are ones to which I will be unable to give a clear answer immediately, without creating more confusion. I will think about those questions and answer in the next class. Thoughtful students also come up with a wide range of questions that are beyond what we are in class. You are welcome to ask such questions, but I may postpone the answer to later in the course or ask you to save the question for a more advanced course. This has nothing to do with your intelligence or ability to grasp concepts; rather, it has to do with the sequential nature of statistical learning.
Expectations
This class is demanding. It covers a lot of material at a pace that students describe as “relentless” or “frantic” or (more positively) “high energy.” It requires considerable effort and outside-of-class time. UM policy states that for each credit hour of a class, undergraduates are expected to work three hours – counting class time, lab time, and study time – to achieve an average grade. If we apply that policy to this graduate-level class, that means a work load of 12 hours per week for this class, implying at least 8 hours per week outside of class/lab. Count on it.
(see
Grading
15+15% Homework (five from textbook, five from lab)20+20% Exams 1 & 2
10% Lab quiz
10% Multivariate analysisexam
10% Participation
The homework is a deliberately sizable portion of your grade, as (1) it is in your best interests to do it and keep up, and (2) it helps take some of the stress off the exams, and can help bring up your final grade if you have difficulty with the time pressure of exams. Each of the 50 questionson the textbook is worth 6 points, or 0.3 points of the final grade. Lab homework will be group-based. You are expected to work cooperatively in groups assigned by TAs. All group members are responsible for the quality of the homework. Only one grade will be given to each group. If your group is not working well for you, please talk to me or TAs as soon as possible. Personalities or schedules occasionally cause conflict that is no one’s fault. For all assignments, the penalty for each day of delay (1 minute to 24 hours) is worth 20% of the assignment grade. Lab quiz questions will be distributed before the quiz. You will not have access to quiz data until the quiz takes place. You cannot bring any notes/do files for lab quiz. However, you can search through Google or use the help command of Stata.
A significant proportion of questions in exams will be adapted from textbook homework, textbook examples, and other review exercises. It is of your interest to work on those questions. The key to a decent grade is to show your work, not only the answers. Exams 1 and 2 will be tested twice. Specifically, each student will take the exams individually for 75 minutes on the exam day; then the group will be tested using the same questions for 25 minutes in the following lecture. The score of the individual exam accounts for 60% of your grade and the score of the group exam accounts for 40%. Open-book multivariate analysis exam questions will be distributed by email at 11:15 am on the examination day. Make sure to check your umn email. If you have any concern regarding exams, come to me before exams. No excuse will be accepted after the exams. Doctor’s note is required for make-up of the exams.
To avoid free-ride, group members will evaluate your participation in group discussion and assignments. The grade of group participation will be based on two confidential group evaluations (one at the 7th week and the other at the 14th week).
Stata Tutorials
Princeton provides free tutorials via the internet
Course Policies
Academic Dishonesty: Students are expected to do their own assigned work. If it is determined that a student has engaged in any form of Academic Dishonesty, he or she may be given an "F" or an "N" for the course, and may face additional sanctions from the University. Academic dishonesty in any portion of the academic work for a course shall be grounds for awarding a grade of F or N for the entire course. See
Diversity and Collegiality: This course draws graduate students from a variety of disciplines. This diversity of academic experience, assumptions regarding learning, and ways of approaching problems is one of the most enriching aspects of the course. In addition, every class is influenced by the fact that students come from widely diverse ethnic and cultural backgrounds and hold different values. Because a key to optimal learning and successful teaching is to hear, analyze, and draw from a diversity of views, the instructors expect collegial and respectful dialogue across disciplinary, cultural, and personal boundaries.
Student Conduct: Instructors are responsible for maintaining order and a positive learning environment in the classroom. Students whose behavior is disruptive either to the instructor or to other students will be asked to leave. Students whose behavior suggests the need for counseling or other assistance may be referred to their college office or University Counseling and Consulting Services. Students whose behavior may violate the University Student Conduct Code may be referred to the Office of Student Judicial Affairs.
Sexual Harassment: University policy prohibits sexual harassment as defined in the University Policy Statement ( Complaints about sexual harassment should be reported to the University according to the website.
Accommodations for Students with Disabilities: Participants with special needs are strongly encouraged to talk to the instructors as soon as possible to gain maximum access to course information. All discussions will remain confidential. University policy is to provide, on a flexible and individualized basis, reasonable accommodations to students who have documented disability conditions (e.g., physical, learning, psychiatric, vision, hearing, or systemic) that may affect their ability to participate in course activities or to meet course requirements. Students with disabilities are encouraged to contact Disability Services and their instructors to discuss their individual needs for accommodations. Disability Services is located in Suite180 McNamara Alumni Center, 200 Oak Street. Staff can be reached at by calling 612/626-1333 (voice or TTY).
Student Mental Health:As a student you may experience a range of issues that can cause barriers to learning, such as strained relationships, increased anxiety, alcohol/drug problems, feeling down, difficulty concentrating and/or lack of motivation. These mental health concerns or stressful events may lead to diminished academic performance or reduce a student's ability to participate in daily activities. University of Minnesota services are available to assist you with addressing these and other concerns you may be experiencing. You can learn more about the broad range of confidential mental health services available on campus via
Acknowledgements
Some sections/sentences were adapted from the syllabus of Dr. Mokhtarian of UCDavis and of Dr. Levison of the Humphrey School.
Lecture and Lab Schedule
WEEK 1
September 5: Introduction
Introduction to course and LU-T data
September 7: Histograms
Read Freedman et al, 3.1-3.3
Lab 1: Introduction to STATA and land use-transportation data
WEEK 2
September 12: Types of variables, average, and standard deviation
Read hand-out on scale of variables and Freedman et al, 3.4, 3.9; Chapter 4
September 14: Uses of the normal curve
Read Freedman et al, Chapter 5
Lab 2: Histograms, mean, median, mode, range, and standard deviation
WEEK 3
September 19: Percentiles and inequality
Read Freedman et al, Chapter 5
September 21: Measurement error, scatter diagrams and correlation coefficient
Read Freedman et al, Chapters 6, 7, 8.1, 8.2, 8.4, 8.6, and 9.1
Lab 3: Review exercises: Chapters 3, 4, and 5
WEEK 4
September 26: Simple regression
Read Freedman et al, Chapter 12.1
September 28: OLS and the r.m.s. error for regression
Read Freedman et al, Chapter 11.1, 11.2 and hand-out on OLS (equations are optional)
Lab 4: Review exercises: Chapters 6, 8, 9, 11, and 12
WEEK 5
October 3: Regression diagnostics
Read Freedman et al, Chapter 11.3-11.5 and handout on diagnostics
October 5: Experiments and observational studies
Read Utts Chapter 5.1, 5.2, 5.4 and Freedman et al, Chapters 1.1, 1.2, 1.4 and 2
Lab 5: Scatter plots and correlation coefficients
WEEK 6
October 10: Probability
Read Dr. Levison’s summary and Freedman et al, Chapter 13 (not including 13.5) and 14
October 12: Exam 1 (covering Weeks 1-5)
Lab 6: Linear regression and diagnostics
WEEK 7
October 17: Binomial Formula, Law of averages, box models, expected value, standard error,
Read Freedman et al, Chapters 15, 16, and 17
October 19: Central Limit Theorem, use normal curve
Read Freedman et al, Chapters 17 and 18
Labs 7 & 8: Review exercises: Chapters 13-18
WEEK 8
October 24: sample surveys and survey methods and chance errors in sampling
Read Freedman et al, Chapters 19 and 20, and Utts, 4.2, 4.4-4.6
October 26: Accuracy of percentages, confidence intervals and accuracy of sample averages,
Read Freedman et al, Chapters21 and 23, and web links on margin of error
Labs 7 & 8: Review exercises: Chapters 13-18
WEEK 9
October 31: Current Population Survey and how to conduct a poll
Read web links and Freedman et al, Chapter 22 (pp. 395--405, 407--408).
November 2: Null & alternative hypotheses, Z- and t-tests of significance
Read Freedman et al, Chapter 26
Lab 9: Review exercises: Chapters 20-23
WEEK 10
November 7: Significance tests for differences in averages
Freedman et al, 27.1, 27.2, 27.5, 27.7
November 9: Chi-square test
Read Freedman et al, 28.1, 28.2, 28.4-28.6
Lab 10: One sample test, independent sample test, paired sample test
WEEK 11
November 14: Multivariate OLS regression
Read hand-out: Ritter, Joseph (2010) “Introduction to Multivariate Regression'' Sections 1-3
November 16: Multivariate OLS
Read Ritter (2010) Section 4
Lab 11: Chi-square test and review exercises: Chapter 26-28
WEEK 12
November 21: Multivariate OLS
Read Ritter (2010) Sections 4 and 5, Hersch and Straton (1995), Multicollinearity
November 23: Thanksgiving
WEEK 13
November 28: Exam 2 (covering Weeks 6-10)
November 30: Multivariate OLS
Read Ritter (2010) Sections 6, 7 and 11
Lab 13: Multiple regression
WEEK 14
December 5: Multivariate OLS
Read Ritter (2010) Sections 6, 7 and 11
December 7: What educated citizens should know about statistics and probability
Read Freedman et al, Chapter 29, Utts (2003), Ziliak and McCloskey (2004)
Lab 14: Lab quiz
WEEK 15
December 12: Multivariate analysis exam
Important Dates:
Students are responsible for all course requirements, including deadlines and examinations. Solutions to lecture homework are available on the Moodle site.
Items / Content / Due DayLab Assignment 1 / Check course website / Sept. 19/21 in lab
Homework 1 / Chapter 3: 8.2, 8.4 on pp. 50-52
Chapter 4: 8.1, 8.6, 8.7, 8.9 on pp. 74-75
Chapter 5: 7.1, 7.7, 7.9 on pp. 93-95 / Sept. 28in class
Lab Assignment 2 / Check course website / Oct. 3/5 in lab
Homework 2* / Chapter 2: 6.1, 6.4, 6.10 on pp. 24-27
Chapter 6: 5.2, 5.4 on pp. 104
Chapter 8: 5.1, 5.7, 5.9 on pp. 134-137
Chapter 9: Exercise Set A 6 on p. 143
Chapter 11: 6.4, 6.5 on pp. 198-199
Chapter 12: 4.1, 4.3, 4.7, 4.8 on pp. 213-215 / Oct. 12in class
Exam 1 / Oct. 12/17 in class
Lab Assignment 3 / Check course website / Oct. 17/19 in lab
Homework 3 / Chapter 13: 6.4, 6.9 on p. 235
Chapter 14: 6.5, 6.7, 6.9 on p. 253
Chapter 15: 3.3, 3.8 on pp. 261-262
Chapter 16: 5.4, 5.7 on p. 285-286
Chapter 17: 6.1, 6.2 on p. 304
Chapter 18: 7.2, 7.11 on p. 329 / Nov. 2in class
Homework 4 / Chapter 20: 6.3, 6.4, 6.7 on pp. 371-372
Chapter 21: 6.5 on pp. 392
Chapter 23: 5.3, 5.4, 5.10 on p. 426-427 / Nov. 9in class
Lab Assignment 4 / Check course website / Nov. 14/16 in lab
Homework 5*+ / Chapter 26: 7.2, 7.5on pp. 495-497
Chapter 27: 6.5, 6.7 on pp. 518-520
Chapter 28; 5.2, 5.9 on pp. 541-543 / Nov. 21(Tuesday) in class
Lab Assignment 5 / Check course website / Nov. 28/30 in lab
Exam 2 / Nov. 28/30 in class
Lab Quiz / Dec. 5/7 in lab
Open-book exam / Multivariate regression / 11:15-12:30on Dec. 12 by email to
* You are not able to get feedbacks from TAs before exams. Please check the solutions of these questions to make sure you understand how to address them. If you have questions, please visit me or TAs during office hour.
+ when answering these questions, please follow the steps we discussed in the class.
1