Skip to content
 

Book reading at Ann Arbor Meetup on Monday night: Probability and Statistics: a simulation-based introduction

The Talk

I’m going to be previewing the book I’m in the process of writing at the Ann Arbor R meetup on Monday. Here are the details, including the working title:

Probability and Statistics: a simulation-based introduction
Bob Carpenter
Monday, February 18, 2019
Ann Arbor SPARK, 330 East Liberty St, Ann Arbor

I’ve been to a few of their meetings and I really like this meetup group—a lot more statistics depth than you often get (for example, nobody asked me how to shovel their web site into Stan to get business intelligence). There will be a gang (or at least me!) going out for food and drinks afterward.

I’m still not 100% sure about which parts I’m going to talk about, as I’ve already written 100+ pages of it. After some warmup on the basics of Monte Carlo, I’ll probably do a simulation-based demonstration of the central limit theorem, the curse of dimensionality, and some illustration of (anti-)correlation effects on MCMC, as those are nice encapsulated little case studies I can probably get through in an hour.

The Repository

I’m writing it all in bookdown and licensing it all open source. I’ll probably try to find a publisher, but I’m only going to do so if I can keep the pdf free.

I just opened up the GitHub repo so anyone can download and build it:

I’m happy to take suggestions, but please don’t start filing issues on typos, grammar, etc.—I haven’t even spell checked it yet, much less passed it by a real copy editor. When there’s a more stable draft, I’ll put up a pdf.

5 Comments

  1. Shravan Vasishth says:

    Very impressive: I just ran ./build.sh and it just compiled. I’ve almost never managed that with my own papers! Look forward to reading it too.

  2. Eric Novik says:

    I am looking forward to reading it and running the code. Thanks for making it public!

  3. Gigli, from the movie says:

    Based on reading the first couple chapters and skimming through the rest the book looks really good. (I was supposed to dedicate this weekend to my Master’s thesis… well, at least its kind of statsy, so reading more about them can’t hurt).

    I wasn’t able to create the PDF on either Windows (through R studio) or in Linux (via build-pdf.sh). First there was something about a cr in a table (about variances of Bernoulli distributed variables) and after deleting the problematic part (couple of et signs) there’s now an undefined control sequence later in the document. Bah, maybe I should update something to make things work, haven’t done that in a while.

    That being said, the HTML version worked right away. I’ll have to transfer that onto my phone so I can read it while going out today.

    • The issues with the pdf were in the source. bookdown is a lot less picky for html generation than for LaTeX, and I never kept the LaTeX up to date. I really wish bookdown was a properly defined and strict format, with meaningful file-located error messages. Until then, bisection is your best bet.

      I went in and fixed the dozen or so bugs in the LaTex that prevented compiling. It still needs another pass for formatting graphs, etc. They still have that yellow background from the html. As I said, it’s a work in progress.

Leave a Reply