Welcome to Modern Data Analysis Techniques
Team taught by Miguel Morales (Physics) and Bryna Hazelton (eScience), the goal of this class is to give you the tools to be a faster and better researcher. One of the transitions students find difficult is going from traditional coursework where there is an answer available—either as an answer key, on the web, or AI might know the answer—to research where the whole point is that no one knows the answer yet. How do you know you are right? How do you become confident that you have the right answer when no answer key is available? The class is organized around four themes to help you convince yourself that you are right: practical statistics, advanced data visualization, collaborative analysis code, and advanced data analysis practices. Along the way we will talk about when you can and can't use AI/ML, designing analysis plans, graduate error propagation, and how to find hardware systematics and software bugs.
Room+
Everyone learns so much better in person please come to class in person whenever possible. That said, many advanced students need to travel for research, so we will offer zoom and recordings on demand. Send Miguel an email if you want zoom for a class, and this will be the link we use (you must sign in to UW zoom to get into the room).
Office Hours
Miguel Morales, Friday 1:30 (right after class), plus by appointment or opportunity. B464.
Bryna Hazelton, Tuesday 2:00 or by appointment. PAB tower, 6th floor (eScience Institute). email: brynah@uw.edu
Grading
We will discuss on the first day. There will be no exams, with the homework and final project form the basis of your grade.
Syllabus
Themes: Practical Statistics; Data Visualization; Collaborative Analysis; Advanced Data Analysis Practices
Week 1
M: Welcome; course overview (slides, video)
W: What does sigma mean? (slides, video)
F: Statistical building blocks (video)
Homework: Homework #1 (git game)
Week 2
M: Introduction to git & GitHub (slides, video)
W: Visualizations
F: Useful Statistical Distributions (video)
Homework: Homework #2 (stats 1)
Week 3
M: holiday
W: Measuring your background, analysis steps (slides)
F: Visualizations pt 2 (slides, video)
Week 4
W: Plot workshop (video)
F: Jackknife tests (slides, video)
Week 5
M: Final project
W: Software Suggestions (slides, video)
F: Software and git patterns (repo, Joel Grus: I don't like notebooks, slides, video)
Week 6
M: Testing and continuous integration
W: Test thickets
F: Stats: parameters, corner plots
Week 7
M: Holiday
W: Stats: trials factors
F: Stats: Frequentist & Bayesian, upper limits, confidence intervals
Week 8
M: Final Projects:
W: Final Projects:
F: Final Projects:
Week 9
M: Final Projects:
W: Final Projects:
F: Final Projects:
Week 10
M: Final Projects:
W: Final Projects:
F: Final Projects: