Eric Aspengren writes:
I’ve been attempting to teach myself statistics and I’ve recently purchased your book Bayesian Data Analysis. I have a small problem, unfortunately. I tend to need to have things explained to me by an actual, physical person (or a video). I’ve been able to use MIT’s online video courses to help with my learning and was wondering if Columbia may have videos of your lectures available. If not, maybe there is a professor who teaches using your text that might have videos of their lectures available somewhere. I tend to be quite thick-headed when learning math.
I’m utterly fascinated by Bayesian methods. I work in politics and I feel there tends to be a lack of quantitative bases for decision making in this field. Conventional wisdom holds sway in many circles and I’m attempting to change that in my area.
My quick reply: I’d recommend starting with my book with Jennifer before moving on to Bayesian Data Analysis. Beyond this, no, I don’t have any lectures online (except these). Actually, some of the online material on Bayesian statistics doesn’t make me so happy (recall our blog discussion on the relevant Wikipedia articles). So I think you have to be careful what you listen to. Or, to put it another way, there’s probably a lot of good stuff out there, but be careful not to take something seriously, just because someone says it in an authoritative manner.
P.S. Also, I write good books and give good one-hour lectures, but I’m not always so great over a one-semester course. I think you’re better off taking a course out of my book but from a different lecturer who can present it from his or her own perspective.
Andrew, I think this idea is a great one. Specifically for you to do a few lectures, not necessarily a whole semester worth, but enough to walk people through exploring a dataset, setting up and fitting a basic bayesian model, setting up maybe a slightly more advanced multilevel model, and then doing some diagnostics and graphical goodness of fit type stuff. Maybe 4 or 5 one hour lectures.
I know that's a significant time commitment on your part, since 5 one hour lectures on video would easily take 60 hours to prepare, give, edit, whatever. But you DO write good books, and you DO give good 1 hour lectures. I watched that one about multiple comparisons, and even the jokes were good :-). The pair together would do a LOT to improve the availability of statistical information on the internet, and might even sell more copies of your books.
I also should say that the fact that you make so many of your own papers available online is a huge plus too.
For introductory level math and stat, the Annenberg Foundation has done some good work and many libraries have copies. They might be too basic for Eric but they are a nice place to get started.
I completely agree with the earlier comment. I would love for you to put some lectures up online. I've got 8 or 9 courses on methods and statistics from ITunes U but I haven't been able to find anything on Bayes.
I'm only in my second methods course and I've been trying to read as much as possible outside of class. Your book with Jennifer Hill is easily the best thing I've read on the subject but actually hearing someone talk about it would be VERY helpful.
As a result of the link above I watched you UW presentation, which I quite enjoyed. But I wonder about your slide that show the relationship between average income and income growth in the 10th and 90th percentiles in the population. The use of income means shouldn't matter very much for 10th percentile, but how about the 90th? I'd like to see this with median incomes rather than means, since the incomes of to ten percent could have a big effect on mean income.
Fred: Those are 10th and 90th percentiles of income, not of medians or means. They're just straight percentiles, I think.
I am educating myself using Gelman, Hill book and it would be fantastic to have a few videos.
Thanks Andrew.
By the way, i found Josh Teneubaum's videos on somewhat related topics interesting and fascinating and good motivational tools.