API-201

Syllabus for HKS API-201 (Z-Section) Quantitative Analysis & Empirical Methods Fall 2017 Draft Version: July 25, 2017 Preliminary, Subject to Change

Class Time: Tuesdays and Thursdays 08:45–10:00am Class Location: L-140 Review Section: Fridays 11:45-1:00am Review Section Location: L-230 Faculty: Dr. Soroush Saghafian Email: soroush [email protected] Office Hours: TBD Faculty Assistant: Beth Tremblay Email: beth [email protected] Belfer-505 TF: Shiro Kuriwaki Email: [email protected] Office Hours: TBD CA: Andrew Hong Email: andrew [email protected] Office Hours: TBD CA: Kristie Liao Email: kristie [email protected] Office Hours: TBD CA: Andy Sugrue Email: andrew [email protected] Office Hours: TBD

1

Brief Course Description This course introduces students to concepts and techniques essential to the analysis of public policy issues. Provides an introduction to probability, statistics, decision analysis, and operations research emphasizing the ways in which these tools are applied to practical policy questions. Topics include: descriptive statistics, basic probability, conditional probability, Bayes’ rule, decision making under uncertainty, Markov chains, statistical inference, hypothesis testing, and regression analysis, among others. The course also provides students an opportunity to become proficient in the use of computer software widely used in analyzing quantitative data.

Overview of API 201 API-201 introduces a range of quantitative tools that are commonly used to inform public policy issues. It provides an introduction to descriptive statistics, probability theory, statistical inference, decision analysis, and operations research with an emphasis on the ways in which they are applied to practical policy questions. Our goal is that by the end of this course you will be able to: 1. Take a data set and a broad descriptive policy question (such as “what has happened to incomes in the U.S. in the last 30 years?”, or “have U.S. screening policies been effective in reducing the rate of cancer?”), figure out what statistical analysis would be most appropriate to answer the question, conduct such an analysis, identify what are the most salient findings/patterns that emerge from the data, and present the findings in a way that is accessible to policy makers. 2. Become skilled in the use of probability, statistics, and decision analysis tools to better tackle real-world personal and policy decisions involving uncertainty. 3. Critically consume policy studies, academic papers, and official reports in which statistical analysis is used. The course content is divided into five broad units: (1) Descriptive Statistics, (2) Probability, (3) Decision Analysis and Operations Research, (4) Statistical Inference, and (5) Regression. Section Z of the course also provides you with an opportunity to become proficient in the use of Stata (or R, or Mathematica, if you so choose) as tools to analyze quantitative data. API-201 is open to all Kennedy School students and is required for first-year MPPs. Students from other graduate programs may enroll with the permission of the instructor. How does the Z Section differ from the A/C/D Sections? Section Z has a similar syllabus as Sections A, C, and D, but it assumes a greater level of mathematical sophistication. It will rely more on calculus and proceed faster than the other sections, allowing more 2

time for applications, in-depth discussions, and a number of advanced topics. Students in the Z section will be given the option to use several statistical software packages (Stata, R, Mathematica), and are expected to use one. By contrast, Sections A, C, and D will focus on the use of Excel. We encourage students considering Section Z to talk with the instructors to determine which section is a better fit. Students with a weaker analytical background are encouraged to consider A/C/D sections. Please note that to enroll in API-202Z, you will need an A or A- in API-201Z or an A in API-201 A/C/D. This course is required for first-year students in the MPP program. First-year MPP students with prior coursework in statistics can place out of the course entirely by taking an exemption exam at the start of the semester. The only way to exempt from API-201 is to successfully pass the exemption exam. Students not enrolled in the MPP program may be admitted with permission of the relevant instructor. Switching Policy. Students may switch from the Z section to the A/C/D sections. Students who switch by October 2 at 12:00 PM are expected to take the midterm exam in the A/C/D section. Students who switch between October 2 at 12:00 PM and October 16 at 12:00 PM are expected to take the midterm in the Z section (on October 5), and this midterm grade will count as the midterm grade in the A/C/D sections. No switches will be allowed after October 16 at 12:00 PM. The process for officially switching between the Z and A/C/D sections will be detailed at a later date. When switching from Z to the A/C/D section, you will be assigned to your original A/C/D section.

Class Policy 1. Classes will consist of lecturing, active and collaborative learning exercises, and discussions. Class participation is very important for learning the material. Hence, it is important that you (a) participate in the class, and (b) spend enough time to work on the reading material (both prior and after each class). Working on the problem sets will also prepare you for the final exam. Please note that the Z section assignments typically require more time from you than the other sections, and the amount of time required depends on your background. We will use Canvas system to post lecture notes, assignments, etc. You should always be prepared to take organized notes if that helps you learn the material. 2. Assignments are due before the lecture starts (i.e., before 8:45am) on their due dates. Late submitted materials will not be graded, and a “0” will be assigned. Note that late submissions include those submitted on or after 8:45am on the due date: even 1 minute late, is late (No exception!). 3. It is imperative that you come to class in a timely fashion. If you come late you may miss the discussion. Furthermore, it disrupts the lecture and the learning environment to have latecomers to the classroom.

3

4. Use of electronic equipment (such as laptops, tablets, iPods, cell phones, etc.) is not allowed in the classroom unless the instructor is notified in advance that you use such devices to take notes. Please turn off and store all electronic equipment before entering the class room. 5. If you have to miss an exam, you will receive a grade of “0” unless you have a doctor’s documentation of illness. In either situation, there will not be any make-up exam. A card from the health center saying that you visited the health center does not count as a doctor’s report. The instructor needs to see a report that clearly states that you were not in a physical condition to take the exam at the scheduled date/time. In that case, your grade will be re-distributed based on the other items that constitute your grade. 6. You are expected to abide by the university policies on academic honesty and integrity. All course activities, including class meetings, homework assignments, and exams are subject to the HKS Academic Code and Code of Conduct. Violations of these policies will not be tolerated, will be reported to the relevant authorities of HKS, and are subject to severe sanctions up to and including expulsion from the University.

Logistics & Requirements Prerequisites. There are no prerequisites for this course. Some of the problem sets will be time consuming and be quite challenging, particularly if you have never worked with statistics before. Please give yourself ample time to work through them and consider working in a group. Statistical expertise is not a prerequisite, but you will still be expected to come to class ready to learn and challenge yourself. The Z section intentionally challenges students to work harder, and learn more about advanced analytical tools and techniques. This might be hard for some students who have weak analytical background, and they may consider A/C/D section instead. Email. All emails to the Instructor, his assistant, and/or TF/CAs must have [API 201-Z] in the subject line. Failure to place this in the subject line could cause your email not to be read. Email is the convenient way to get in touch with the instructors to get an answer to a short question. However, be aware that the professor, his assistant, and TF/CAs have many other work obligations and probably keep different schedules from you. Therefore, although we do guarantee that we will answer all emails we receive, you should not expect us to answer each of them right away. The most reliable way to get your questions and concerns answered is to attend office hours: keep up with lectures and reading materials, and get started on assignments early, to be prepared to pose questions in office hours. Class Attendance. We expect that you will arrive on time to class and do your best to attend every class. If you need to miss a class due to an emergency, it is your responsibility 4

to obtain missed notes and course announcements from another student; there is no need to email the instructors. All lecture notes will be posted on the course website. It is highly recommended that you go through the lecture notes prior to attending the class. On the sessions that we will work on case studies, you should have studied them in advance and be ready to answer questions and/or discuss the case. Suggested Textbooks. The study of probability and statistics is rapidly moving to online platforms—many of which provide free and excellent overviews of all of the material we cover in this class. In addition, all course lecture notes will be posted to the course website. However, along with API 201 A/B/C, there is one strongly suggested textbook: • Introduction to the Practice of Statistics. Moore, McCabe and Craig. W.H. Freeman, 8th Edition. This text can easily be rented, bought used, or bought new, all online. Unlike students in the A-B-C Sections, there is no need for Z Section students to purchase any of the online supplementary subscriptions. In addition, we strongly recommend the purchase of: • The Cartoon Guide to Statistics. Gonick and Smith. Harper Perennial. The Cartoon Guide provides an intuitive, easy-to-follow overview of most topics we cover in this course. It is strongly recommended, particularly for those new to statistics. For a more rigorous treatment of probability for modeling and analysis, we recommend the following book: • Introduction to Probability Models, Ross, S. M., 10th Edition, Academic Press. We may occasionally use materials from this book. Suggested Fun Readings. For fun readings related to the course material, you may consider reading the following books: • How Not to Be Wrong: The Power of Mathematical Thinking, Ellenberg, J., Penguin Books. • More Damned Lies and Statistics: How Numbers Confuse Public Issues, Best, J., University of California Press. • Statistics Done Wrong: The Woefully Complete Guide, Reinhart, A., William Polllock.

5

Statistical Software Packages. One of the objectives of the course is to help you gain proficiency using statistical software packages. Section Z will use Stata (or R or Mathematica as other options) to conduct statistical analysis throughout the course. Problem sets will contain exercises designed to get you to practice the basics of Stata (or R or Mathematica if you so choose). Tutorials will be offered during the first weeks of class to introduce Stata, and assistance provided throughout the semester by the teaching fellow and course assistants. Only Stata will be officially supported by the teaching fellow and course assistants, although the teaching fellow can answer questions about the use of R or Mathematica. Students concerned about or unable to invest the time and effort needed to learn a statistical software language (such as Stata or R or Mathematica) should consider opting into the A/B/C Sections instead.

Grading Your grade in this class will be composed of • 15% - Problem sets • 10% - Class participation and engagement – see below • 15% - Final Exercise • 15% - Midterm #1 • 15% - Midterm #2 • 30% - Final exam Final letter grades will be determined on a curve using the Dean’s Recommended Grade Distribution. Problem Sets. The best way to learn probability, statistics, and related modeling and analysis is to practice. Thus, problem sets will be assigned almost every week and due at the beginning of class. There will be no credit for late assignments. You should plan to spend approximately 6-8 hours on each problem set, although it may be more depending on your background. Problem sets will be posted on the course website approximately 10 days before they are due. Under the Kennedy School Academic Code, the problem sets for this course are “Type II” assignments. You are encouraged to work in a study group, but must submit your own handor type-written solutions. It is not acceptable to work on one electronic document as a group and submit identical, or nearly identical versions. Examples of assignments that are not in accordance with the HKS academic code include photocopies or reprints of substantially identical assignments, printouts of substantially identical Stata or R or Mathematica tables or graphs, and copies of solutions from previous years. Problem sets are due at the beginning 6

of class on the date they are due. Please indicate on the first page the names of the students you worked with. The lowest problem set grade will be automatically dropped in the calculation of your overall problem set grade. Class Participation& Engagement. Student participation can substantially enrich the learning experience for both the students and the instructor. In this spirit, class participation is considered as part of your final grade (although with a small percentage). Students are also encouraged to go through the lecture notes and other references to prepare before attending each session. Final Exercise. The final exercise is a group-based project that engages all of the skills acquired in the class by asking you to take the statistical skills learned and apply them to the analysis of some original, interesting, or otherwise professionally relevant data set. Groups should be 3–4 students total. The components of the final exercise include (1) a short write-up of your analyses and results, and (2) a short power-point presentation of the key points. The final exercise is due on December 1, 2016. More details and deadlines will be provided later in the course. Students are encouraged to start thinking about their groups and also about potential data sources as early in the term as possible. Each group is essentially responsible for choosing a topic, collecting data, and providing useful analyses. Midterm Exams. There will be two midterms, and a final exam. These will be closed book/notes and a formula sheet will be provided. Calculators may be used, but statistical functions on them may not. Calculators that allow text storage are not permitted. The two midterms will be given in class on Thursday, October 5, and Tuesday, November 14. Please schedule your travel plans accordingly and arrange to be in the classroom on time. Midterm 1 will cover lectures 1-9 and midterm 2 will cover lectures 10-19. Final Exam. Our exam has been scheduled by the HKS Registrar on Friday, December 8 9am to 12pm. We have no discretion to change the time or date of the final exam. Schedule your travel plans accordingly. The final exam may ask questions related to materials covered in any of the lectures from the beginning to the end of the semester. Regrading Policy. Requests for regrades will be accepted only in writing, with a clear statement of what has been mis-graded, and why, and within one week of return of your graded work. Note that we reserve the right to review all of your answers, and you might end up having more points deducted than if you hadn’t requested a regrade. Regrading requests should be submitted directly to the course TF.

7

Tentative Schedule This schedule is subject to change, depending on how far we get in each class and how much discussion the topics generate.

August 31 (Lecture 1): Introduction and Class Overview September 5 (Lecture 2): Descriptive Statistics • MMC 1.3, 2.3, 2.7 September 7 (Lecture 3): Probability I • MMC 4.1, 4.2, 4.5 September 12 (Lecture 4): Probability II • MMC 4.5 • Problem Set #1 Due September 14 (Lecture 5): Conditional Probability and Bayes’ Rule • Lecture Notes, Ross 1.4, 3.1-3.4 (optional) • Mammogram article September 19 (Lecture 6): Application: Public Pensions in Mexico • Case study on Public Pensions in Mexico • Problem Set #2 Due September 21 (Lecture 7): Discrete Distributions • MMC 4.3 (through page 255), 4.4, 5.2 September 26 (Lecture 8): Continuous Distributions • MMC 1.3, 4.3 (from page 255 onward) • Problem Set #3 Due September 28 (Lecture 9): Introduction to Operations Research and Decision Analysis • Lecture Notes 8

October 3 (Lecture 10): Introduction to Stochastic Processes, Markov Chain Analysis, and Queueing Theory • Lecture Notes, Ross 2.9, 4.1 (optional) • Problem Set #4 Due October 5 Midterm 1 (In-Class) October 10 (Lecture 11): Introduction to Inference, the Central Limit Theorem, and the Law of Large Numbers • MMC 3.3, 5.1 • Problem Set #5 Due October 12 (Lecture 12): Inference with Means I • MMC 6.1, 6.2, 7.1 October 17 (Lecture 13): Inference for Means II • MMC 7.2, 8.2, 6.3 • Problem Set #6 Due October 19 (Lecture 14): Inference for Proportions I • MMC 8.1 October 24 (Lecture 15): Inference for Proportions II • MMC 8.2, 6.4 • Problem Set #7 Due October 26 (Lecture 16): Paired Data • MMC 7.1, 8.2 (review) October 31 (Lecture 17): One-ANOVA • MMC 12.1, 12.2 • Problem Set #8 Due 9

November 2 (Lecture 18): Application: Oregon Health Care Experiment • Oregon Health Care experiment case study November 7 (Lecture 19): Chi-Square Tests • MMC 9.1, 9.2, 9.3 • Problem Set November 9 (Lecture 20): Regression Analysis I • MMC 2.3, 2.4, 10.1 (p. 563–580) • Problem Set #9 Due November 14 : Midterm 2 (In-Class) November 16 (Lecture 21): Regression Analysis II • MMC 11.1, 11.2 November 21 (Lecture 22): Regression Analysis III • Lecture Notes • Problem Set #10 Due November 23: THANKSGIVING HOLIDAY RECESS – NO CLASS November 28: Final Presentations November 30: Final Exam Review • Final Exercises Due December 8: Final Exam [9:00-12:00]

10