Syllabus: Stat 370 Spring 2017

From Sean_Carver
Revision as of 07:48, 15 January 2017 by Carver (talk | contribs)
Jump to: navigation, search

Introduction to Statistical Computing and Modeling (STAT 370) Spring 2017 Section 001 [UNDER CONSTRUCTION]

Instructor: Sean Carver, Ph.D., Professorial Lecturer, American University.

Contact:

  • office location: 107 Gray Hall
  • email: carver@american.edu
  • office phone: 202-885-6629

Course Description (from department website): The basics of programming using the open source statistical program R. Data analysis, both numerical and qualitative, including graphical and formal inference. Applications include numerical methods, text mining, modeling, and simulation. Usually offered every spring. Prerequisite:

Prerequisite: MATH-221 and STAT-202 or STAT-203, or permission of instructor.

Text: The Book of R, by Tilman M. Davies, No Starch Press, 2016.

Optional Text: Analyzing Baseball Data with R, by Max Marchi and Jim Albert, CRC Press, 2013. This is a great book, if you like Baseball, data science, or especially both. We are going to use it to learn how to simulate a baseball game with a Markov Chain. I may be able to legally provide the relevant chapter of the book, if you want to save money. Don't like baseball, or don't know the rules? Don't worry, neither do I, but this example is fantastic from a pedagogical perspective, and I think you will agree, regardless of your interest and knowledge in the sport.

Software: Please install the following software on the machines you intend to use for this class: R, RStudio, Git, XQuartz (Mac), GitBash (Windows). You are welcome to use a lab computer or your laptop during class.

Learning Outcomes: Students will be able to

  • Submit work as a dynamic document via GitHub
  • Use R as a powerful calculator.
  • Write basic programs using control and data structures.
  • Import data from external sources.
  • Perform analyses using regression, text mining, and simulation.
  • Use SQL databases to retrieve data with specified features.

Office Hours: Students are strongly encouraged to come to office hours if they need or want help.

My office is Gray Hall, Room 107. Office hours are TENTATIVELY scheduled as follows: (may be adjusted throughout the semester)

  • Tuesday, Wednesday, Friday: 4:00 PM TO 6:00 PM.

NOTE: If you would like to come to office hours on a regular or irregular basis and you have a compelling reason why you cannot make it during the hours listed above, please send me an email. I cannot guarantee that I will be able to find a time that works (this semester will be a very busy one for me), but I will try.

Class times and locations:

  • Tuesday, Friday 11:20 AM TO 12:35 PM, ANDERSON B-13

Important Dates:

  • January 17 (Tuesday): First day of class.
  • January 20 (Friday): Inaugaration Day, no class.
  • January 24 (Tuesday): Initial Project Brainstorm (come with ideas).
  • February 14 (Tuesday): Project Proposals due.
  • March 12 - 19: Spring Break, no class.
  • March 21 (Tuesday): Project Updates due.
  • March 31 (Friday): Midterm Exam.
  • April 28 (Friday): Last day of classes and final projects due.
  • May 12 (Friday): Grades due to registrar, no final exam.

Projects: In my experience, there is no better way to learn to code than to engage with a project that you feel passionate about. For the first month of class, we are going to some spend class time finding and defining projects that meet the course objectives, and that inspire this kind of passion in us. I anticipate that some of these projects may involve a lot of work for one person, which is why I am going to teach some of the tools that open-source software engineers use to facilitate collaboration. Collaboration will be accomplished through the cloud based service GitHub, and a local program called Git. As a pedagogical exercise, you will use these tools to collaborate on writing a children's story in R-Markdown (available within R). After the exercise you will not be compelled to collaborate, but the option will be there. Do group projects give you nightmares? The open source community has figured out paradigms for successful collaboration, although they are not widely used outside of the coding community because they especially simple to learn. These tools make attribution for work very transparent. This transparency will make it possible for students to get credit for contributions to several projects, not just one. You will be graded on the body of your work, if you choose to divide your effort among more than one project. If you start a project, it will be up to you whether you want to allow and invite others to contribute. I should warn the class, however, that I will be traversing the collaboration learning curves at the same time I teach the material, so bear with me. Project milestone will be a proposal (February 14), and project update (March 21), final submission (April 28, last day of class). Each project will have a cloud based repository. GitHub is best for collaboration, but they charge for private repositories -- GitLab doesn't charge for private repositories. Either way, you will make it possible for me to pull the most recent version of the project onto my computer for grading. I will do this at a specified date and time -- when it is due. For collaborative projects you will also be turning in a portfolio of your submissions which should be easy to generate.

Homework: For homework we will use private repositories on GitLab, except for the children's story assignment. For the children's story, each student will start their own GitHub repository. Both the private homework repository and the public children's story repository will be separate from the project repositories. (Project repositories will be either public or private depending on the decision of the project leader. There will be work to do most classes. You will commit this work to the cloud whenever you want. You can update the work as you progress. At specified times and dates, I will pull your work from the cloud onto my computer to grade it. The specified time for the pull will be when it is due. I won't see any changes you commit after the pull time..

Midterm: The midterm project will be a coding exercise, completed in class on March 31. You will have access to your books, notes, and Google, but you will not be able to interact with another live person. Pull time will be the end of class on that day. Midterm examinations will be pulled from the private homework repository.


Tentative grading scheme:

ITEM PERCENT
Attendance and Participation 15%
Homework 10%
Midterm 25%
Project 25%

Class Etiquette: Please give the class your full attention and refrain from talking during lectures, texting, surfing the web, and similar distractions. If you need to attend to something urgently, it is OK to excuse yourself from the classroom.

Please participate in class by asking questions when you do not understand something. Invariably other students benefit from these questions. Please engage in discussions, and please engage with the class, generally.

Academic Integrity: Cheating is not acceptable and will not be tolerated. Consider this: in subtle ways, cheating to get a better grade on an exam can result in lowering the grades of some of your classmates. Certainly this is true when a specific curve is used to assign grades. Even when I don't use curves explicitly, they can be implicit in decisions about writing and grading exams. As required by the policy of American University, I will report all suspected cases of cheating to the Dean's office who will proceed to investigate and adjudicate the issues. Cheating is giving or receiving unauthorized assistance on exams, from other students or other people, from notes, from books, or from the web. When inappropriate copying between students is caught, both parties may be culpable.

Homework, Attendance and Participation Policy: Homework is worth 10% of your grade, with each assignment weighted equally (10 points maximum). The due date for homework is technically the end of class, one week after it is assigned, however you should make every effort to complete each assignment as soon as possible. That said, you can turn in homework anytime, up to Wednesday, December 7. Homework turned in after the technical due date will receive a maximum of only half credit (5 points). I plan to post deadlines on this website.

I like to give the solutions to homework problems at the same time I assign the problems. Conscientious students, who wrestle with problems before looking at the answers, benefit from having instant feedback about their solutions, right, wrong, or incomplete. Less conscientious students who use the answers to easily complete the assignments often do poorly on exams. The responsibility for your education rests in your own hands. Don't be one of the outliers who use shortcuts to avoid preparing for the exams. Concerning homework, you are encouraged to work with your classmates, if you find that helpful. In fact, you are encouraged to do whatever you find most helpful with the homework, but by turning in a solution to a problem, you pledge that you understand the solution, or that you talked to me in office hours or during or after class and made a good faith effort to understand how to do the problem. If it looks like you got the full benefit from the assignment, and if you turn it in by the due date, I will award you a perfect 10 points. I may mark you down if it seems that you have copied the answers without including any of the required calculations. You must include your work.

Additionally, 15% of your grade is for class attendance and participation. I understand that there are times when you cannot make it to class for compelling reasons. To accommodate the unavoidable, I will forgive a fixed number of absences for everyone when I compute the final grades. If you need to miss more than several classes, please see the Dean of Students. I will only erase absences from my grade book with a request through the Dean of Students. That said, I do appreciate an email when you can't make it to class. Absences on exam days must be excused through the Dean of Students.

I plan to post grades to Blackboard, promptly.

Public Service Announcement: A representative of AU's Students Against Sexual Violence (SASV) approached me and asked me to include on my syllabi a list of resources available for survivors of sexual assault and their friends. While sexual violence is by no means the only challenge faced by students, I agree that this issue merits particular attention, so I am honoring her request by attaching the list she gave me:

Sexual Assault Resources

  • It’s never the survivor’s fault. There are many people you can talk to if you or someone you care about has been sexually assaulted:
  • AU's Sexual Assault Prevention Coordinator Daniel Rappaport (rappapor@american.edu)
  • AU's Coordinator for Victim Advocacy Sara Yzaguirre (sarayza@american.edu)
  • DC SANE Program (Sexual Assault Nurse Examiner) 1-800-641-4028
  • The only hospital in DC area that gives Physical Evidence Recover Kits (rape kits) is Medstar Washington Hospital
  • DC Rape Crisis Center: 202-333-7273
  • Students found responsible for sexual misconduct can be sanctioned with penalties that include suspension or expulsion from American University, and they may be subject to criminal charges
  • If you want to submit a formal complaint against someone who has sexually assaulted you, harassed you, or discriminated against you based on your gender identity or sexual orientation, you can do so online at http://www.american.edu/ocl/dos/, or contact the Dean of Students at dos@american.edu or 202-885-3300. These are Title IX violations, and universities are legally required to prohibit these actions.
  • Resources on campus that are required to keep what you tell them confidential are Daniel Rappaport, Sara Yzaguirre, ordained chaplains in Kay, and counselors at the counseling center. (OASIS may also belong here but it didn't exist when this list was created.)