Stat 6103, Bayesian Data Analysis, TuTh 1-2:30 in room 428 Pupin Hall:

We’ll be going through the book, section by section. Follow the link to see slides and lecture notes from when I taught this course a couple years ago. This course has a serious workload: each week we have three homework problems, one theoretical, one computational, and one applied.

Stat 6191, Statistical Communication and Graphics, TuTh 10-11:30 in room C05 Social Work Bldg:

This is an entirely new course that will be structured around student participation. I’m still working out the details but here’s the current plan of topics for the 13 weeks:

1. Introducing yourself and telling a story

2. Introducing the course

3. Presenting and improving graphs

4. Graphing data

5. Graphing models

6. Dynamic graphics

7. Programming

8. Writing

9. Giving a research presentation

10. Collaboration and consulting

11. Teaching a class

12-13. Student projects

**Why am I teaching these courses?**

The motivation for the Bayesian Data Analysis class is obvious. There’s a continuing demand for this course, and rightly so, as Bayesian methods are increasingly powerful for a wide range of applications. Now that our book is available, I see the BDA course as having three roles: (1) the lectures serve as a guide to the book, we talk through each section and point to tricky points and further research; (2) the regular schedule of homework assignments gives students a lot of practice applying and thinking about Bayesian methods; and (3) students get feedback from the instructor, teaching assistant, and others in the class.

The idea of the communication and graphics class is that statistics is all about communication to oneself and to others. I used to teach a class on teaching statistics but then I realized that classroom teaching is just one of many communication tasks, along with writing, graphics, programming, and various forms of informal contact. I think it’s important for this class to *not* be conducted through lectures, or guest lectures, or whatever, but rather as much as possible via student participation.

I wish I could take the second course in particular.

Indeed, that 2nd course looks terrific.

When I was teaching (computer science), we often said that if we got the problems right, students would learn a lot regardless of what we said in lectures… So I’ll be interested to see the nature of the assignments for each week.

Will you be posting the material from the 2nd class? I’d be interested in following along as much as my work schedule provides.

Will you be posting a video like Andrew Ng does for his machine learning class?

Andrew, please let us know how the Stat 6191 class worked out once you’ve finished it. There are lots of people (like me!) thinking through how to incorporate more communication & visualization skills for graduate students.

[…] here I am preparing my course on statistical computing and graphics and thinking of points to mention during the week on […]

Looks like a fun course on Statistical Communication on Graphics. Why are you only covering Dynamic Graphics and not Interactive Graphics as well?

Antony:

I don’t really know the difference between dynamic graphics and interactive graphics. During that week of the class I’d like to give students practice in both. Do you have suggested readings and assignments for them, something do-able that would give them a bit of experience?

Andrew, if they will have some familiarity with R you could have them try rCharts (https://github.com/ramnathv/rCharts). Otherwise they could have a quick exercise making an interactive from Google Spreadsheets. The thing about interactive graphics is that you/they need an easy interface from their stats software to JavaScript, which they won’t want to learn. If you want some examples you could just point them at the NYT Upshot page and then let them design one on paper. And by the way I would heartily recommend Isabel Meirelles’s book “Design for information” to introduce ideas from psychology and design as to why some things work and others don’t.

Andrew:

An oversimplified answer would be that Dynamic Graphics emphasise movement (animation, rotation, brushing) and Interactive Graphics emphasise interaction (querying, selectign, linking, reformatting). Theus and Urbanek “Interactive Graphics for Data Analysis” is a good text, getting the students to try out Martin Theus’s Mondrian would be even better. Learning to use Interactive Graphics is like learning to ride a bicycle: a textbook or instruction manual is a poor substitute for getting someone who can do it to show you how.

For interactive graphics I’d emphasize some of the canonical techniques: brushing and linking and dynamic querying/direct manipulation. For brushing and linking there are more recent systems like ggobi, though Wills 1995 paper on Visual Exploration of Large Structured Data Sets is a nice paper for students to read to get the gist of the technique. Much of Ben Shneiderman’s early work laid the foundation for interacting with graphs through direct manipulation and dynamic querying — see his early systems HomeFinder and FilmFinder. The TimeSearcher system is also nice for visual queries on stock data. There’s of course a lot more, but these techniques are foundational in how I’ve learned/teach interactive graphics.

Thanks for having me join the class today! It was a lot of fun.