Fall 2022: PSTAT 215: Bayesian Data Analysis
In this course, we will explore the data science lifecycle: question formulation, data collection & cleaning, exploratory data analysis & visualization, statistical inference and prediction, and decision-making.
This document and others linked within it should be your PRIMARY source for understanding the expectations of this course. Be sure to read it carefully. You must contact the instructor for clarification if you receive information from any other source that is in contradiction to what is provided below.
Below are the links to different sections of the syllabus:
COURSE INFORMATION
- Instructor: Alex Franks
- Lecture: Tuesdays and Thursdays, 9:30am-10:45pm
Prerequisites:
Programming experience: Intermediate familiarity with R and Rmarkdown is required. Familiarity with ggplot and the tidyverse a plus.
Course Topics:
At the end of the course, a successful student will be able to build and refine statistical models using the Bayesian paradigm and utilize Monte Carlo methods for statistical inference. Topics include:
- One parameter models
- Hierarchical models
- Monte Carlo computation
- Markov chain Monte Carlo
- An introduction to probabilistic programming
Textbook
- Required: Andrew Gelman, John Carlin et al. Bayesian Data Analysis; http://www.stat.columbia.edu/~gelman/book/
Additional optional resources:
- Optional: Christian Robert The Bayesian Choice; [https://link.springer.com/book/10.1007/0-387-71599-1]
- Optional: Richard McElreath Statistical Rethinking; https://xcelab.net/rm/statistical-rethinking/
- Optional: Jim Albert Bayesian Computation with R https://www.springer.com/us/book/9780387922973
- Optional Alicia A. Johnson, Miles Q. Ott, Mine Dogucu Bayes Rules https://www.bayesrulesbook.com/
Computing Platform
The computing platform (Jupyter Notebooks) for the course is hosted at https://pstat215.lsit.ucsb.edu. You can do all of you work here or on your own computer. Since we are using quarto, if you are working on your own computer, you must download the newest version of Rstudio.
If you are working on the lsit site, you should use the following link to automatically sync course material: https://tinyurl.com/3bs8uu8d
If you are working on your own computer you can get all the courses files, as they are released, via github: https://github.com/ucsb-pstat215/fall22
Contact us on Nectir!
All class related questions should be handled through Nectir. Our class page is here: https://ucsb.nectir.io/group/PSTAT215-F22. An invite link was sent out via Gauchospace or you can use the invite link here: https://app.nectir.io/invite/7hnEew
We ask that when you have a question about the class that might be relevant to other students, post it on Nectir instead of emailing me. That way, everyone is onthe same page and everyone can benefit from the response.
Drop-in Hours
It’s expected that some aspects of the course will take time to master, and the best way to master challenging material is to ask questions.
Drop-in hours will be immediately after class Thursday, 3:30-4:45pm in Prof. Franks’ office, South Hall 5522.
ASSESSMENTS AND GRADES
Your mastery of class material will be assessed in the following ways, and final grades will be computed as follows:
- 5% Class Participation
- 40% Homework Assignments
- 20% Take-home Midterm Exam (November 3, tentative)
- 35% Final Project Writeup Due (Wednesday, March 16, 2022)
Participation
Lecture and section attendance is optional but is highly encouraged. You are adults and are responsible for your learning. However, everybody benefits when there is more participation and engagement with the material during lab and lecture.
The participation portion of your grade will also include providing good answers on Piazza and engaging with the various activities that the instructor will provide throughout the quarter.
Assignments
Weekly homework assignments are a required part of the course. You are allowed to work with a partner, but everybody most turn in their own work. Include your partners name on your work.
All assignments must be an Rmarkdown (.rmd) or Quarto (.qmd) document.
Midterm Exam
Take Home Exam, Thursday November, 3rd.
Final Project
Due on the final exam day, more information to come.
Late Policy
Homework will be accepted up to 1 day (24 hours) late; a homework submitted within 24 hours after the deadline will receive a 10 point deduction. No assignments will be accepted 24 hours after the deadline.
If there is a properly documented family emergency, extended illness, documented required court appearance, or other situation beyond the students’ control (with appropriate official detailed documentation) the instructor may extend an assignment deadline, entirely at the instructor’s discretion.
Learning Cooperatively
We encourage you to discuss all of the course activities with your friends and classmates as you are working on them, either on Nectir, or through a personally chat or zoom.
Academic Honesty
Cooperation has a limit. You should not share your code or answers directly with other students. Doing so doesn’t help them; it just sets them up for trouble on exams. Feel free to discuss the problems with others beforehand, but not the solutions. Please complete your own work and keep it to yourself. The exception to this rule is that you can share everything related to a project with your project partner and turn in one project between you.
Penalties for cheating are severe — they range from a zero grade for the assignment up to dismissal from the University, for a second offense.
Rather than copying someone else’s work, ask for help. You are not alone in this course! We are here to help you succeed. If you invest the time to learn the material and complete the projects, you won’t need to copy any answers.
Slides and Recordings
All lecture material including slides will be posted after class on the cloud server. Recordings of any remote lectures will also be posted. Lecturers will be in person and not recorded, unless stated otherwise.