Skip to content

Ben Lambert. 2018. A Student’s Guide to Bayesian Statistics.

Ben Goodrich, in a Stan forums survey of Stan video lectures, points us to the following book, which introduces Bayes, HMC, and Stan:

If Ben Goodrich is recommending it, it’s bound to be good. Amazon reviewers seem to really like it, too. You may remember Ben Lambert as the one who finally worked out the bugs in our HMM code for Stan for animal movement models; I blogged about it a couple years ago and linked the forum discussions where it was being worked out.

The linked page has answers to the exercises and an associated Shiny app for exploring distributions. There are also videos for a course based on the book:

I haven’t seen a copy, but I am very curious about the section titled “Bob’s bees in a house”, as it’s an example I’ve used in courses. I didn’t come up with the analogy—I borrowed it from a physics presentation on equilibrium in gases or something like that I’d seen somewhere.

Does anyone know if the Kindle version of this book is readable? Living and working in NYC, I have very limited space for physical books.


  1. Eric Novik says:

    I have the Kindle version and it is fine. I mean the typesetting is not great as you can imagine but you could read the formulas and see the graphs. It would probably be best to read it on a computer using Kindle software or iPad or something like that. On the Kindle device, you have to squint to see some of the formulas.

  2. Rok Č. says:

    I am reading it on the Kindle Reader for the iPad and the formulas are readable.

  3. When I finish with the three I’m reading now, I would like to go through that. Thanks for recommendation

  4. Jonathan says:

    The text is fine. Magnifying some of the graphics is challenging but doable. I’ve enjoyed it.

  5. I read the physical book first and then bought the kindle edition (iPad) for when I am on the road. It’s much better formatting than e.g., the MCMC textbook by Gilks et al, which was all but unreadable.

    I really like this book and plan to use it as a textbook the next time I teach Bayes. It’s detailed without being too mathematical. It’s going to be hard to beat this book as an introduction.

    It is also very very funny in a British understated kind of way. Example:

    “Markov was alive during a time of great political upheaval and was not afraid to give his opinion on matters that he deemed important. In 1913 the Romanov dynasty celebrated 300 years of rule (which, in hindsight, probably was not a particularly sensible thing to celebrate at the time). Markov showed his disapproval by organising a celebration of his own – a celebration of 200 years since the Law of Large Numbers was proved! Although it was likely not such a debauched affair, we nonetheless appreciate its sentiment.

    Lambert, Ben. A Students Guide to Bayesian Statistics (Kindle Locations 5339-5343). SAGE Publications. Kindle Edition. “

  6. Dustin says:

    Generally I love his videos. Have used them to fill in my knowledge gaps on the fly when I need to use a new method for something.

  7. Andrew says:

    I’m so so glad to hear about this book, and also McElreath’s book. When John Carlin and I wrote our BDA book, back in the early 1990s, it was because there was no existing book on applied Bayesian statistics. There were a few books out there, but all of them were theoretical, and focused on some mixture of algebra, theory, and very simple examples. It’s good to see new books coming out, at all levels, with an applied perspective and a “Bayesian workflow” approach.

    • Michael Schwartz says:

      Thanks for the link to this great information.

      I also want to put in a (re-) plug for Richard McElreath’s book, and youtube lecture series. I personally found these to be incredibly useful, and tremendously entertaining. A rare combination for stats education (though, seemingly, less rare in the Bayesian world – someone should analyze that).

  8. Eddie says:

    I bought and worked through this book about a month after it came out last April. I love Lambert’s videos for both probability concepts and, especially, econometrics. The exposition in the book is okay, as others have noted above, and certainly it’s one of the better descriptions of MCMC (particularly, Hamiltonian Monte Carlo) that I have read.

    That said, the latter sections on hierarchical modeling and its applications are really lacking. The real Achilles heel of Lambert’s book, though, is his end-of-chapter exercises. These are totally out-of-line with the book’s stated context of a student’s introduction of Bayesian methods and serve more to confuse and frustrate rather than illuminate. They look more like problems in BDA 3 than they do illustrative examples for individuals encountering this material for the first time.

    So, it’s worth reading Lambert’s sections on MCMC/HMC, but the payoffs from the rest of the book, and especially the exercises, are not worth the time.

  9. jrkrideau says:

    As a definite non-statistician, I have gotten through the first four videos and I’d say they are verging on brilliant. Simple enough that I could probably recommend them to an historian friend whose last math class was probably 30 years ago but who wants to learn some basic stats.

    I am definitely finding them a great review of things I (think) I knew.

    It hurts but it looks like I’m going to have to break down and buy the book.

  10. Alex says:

    I’ve become pretty sensitive to book prices, so I hope this is a pretty fantastic book if it’s $100+. Ouch!

Leave a Reply