Skip to content

He’s looking for a Bayesian book

Michael Lewis wrote:

I’m teaching a course on Bayesian statistics this fall. I’d love to use your book but think it might be too difficult for the, mainly, graduate social work, sociology, and psychology students likely to enroll. What do you think?

In response, I pointed to these two books that are more accessible than mine:

Statistical Rethinking

A Student’s Guide to Bayesian Statistics.

Also, Regression and Other Stories, but that’s not really a Bayesian book, even though it has some Bayesian stuff in it.


  1. awinter says:

    Statistical Rethinking is what I am using in our courses and workshops for folks. I like it. Most folks are from non-stat/mathy backgrounds. I also borrow from but use rstanarm for the examples.

  2. Madeleine says:

    Is “Regression and Other Stories” available yet?

  3. Michael Schwartz says:

    Another +1 for rethinking. It’s excellent for non-statisticians. Also, there is a great youtube series (actually, three of them) from McElreath’s teaching of the course. This would be a terrific supplement for students.

  4. Dave C. says:

    How about William Bolstad’s “Introduction to Bayesian Statistics” for a basic level?

  5. Shravan says:

    What is the background of the intended audience? Impossible to suggest anything without that information.

  6. jrkrideau says:

    I am slowly working my way through Lambert’s A Student’s Guide to Bayesian Statistics. on my own and finding it pretty good. Overall, I get the feeling that the overall math level is not too high for psychology student speaking as a former psychology grad student whose last calculus course was over 40 years ago and has no background in Bayesian statistics.

    Lambert has extensive on–line resources. I have only viewed some of the videos and they seem good. Overall, with a good instructor, I think it would be a good text.

    As an aside, I was reading the book while in hospital: Doctors and nurses recoiled in shock (horror?) when they saw the open book.

    • Eddie says:

      Lambert’s book is okay for expositional purposes, but his end-of-chapter problems are way too far afield for an early non-stats graduate student. Also, the later chapters are not nearly as well developed as the first half of the text. Lambert’s videos are, admittedly, excellent, however and the book has parts where it’s language shines through. But for practical implementation of the methods definitely look elsewhere.

  7. Alex says:

    I’ve liked Kruschke’s Doing Bayesian Data Analysis. Kruschke is a psychologist so it’s definitely written from a practical approach over a math approach, although there is still some math in there. It’s written mostly for JAGS but there is some RStan as well.

  8. Adam H says:

    Another +1 for rethinking. It’s the only stats book I’ve ever read for fun. Accessible and informative for a wide variety of audiences.

  9. Andrew Deece says:

    Perhaps a good place to start are
    (1). Statistics & Data Analysis for Financial Engineering, with R examples. 2nd Ed., David Ruppert & David S. Matterson
    Chapter 20, Bayesian Data Analysis & MCMC

    (2). Information Theory, Inference & Learning Algorithms, David J.C. MacKay,
    Chapter 37, Bayesian Inference & Sampling

    (I’ve got a small slide pack, which I’m happy to share with you. If you’re interested, just let me know).

Leave a Reply to Sean S