Workshop on Bayesian Data Analysis (Part 3)

This workshop will be the third in a series introducing the concepts of Bayesian Data Analysis. It is expected that this will be the finish off the series.

As before, I strongly recommend that attendees bring their own laptop as the material is based around interactively working with the problems, running code yourself, and trying to create plots and charts to visualise the output.

We continue the analysis of a coin toss problem, and previous workshops have laid the foundation of understanding conditional probability, and how the Bayesian approach uses this concept to help model and understand the problem.

In this workshop we will introduce the hierarchical approach, how JAGS allows us to use these hierarchical models, and use the outputs of these samplers to make inferences on the various parameters of the models. We will focus in particular on using visualisation of the data, and so a basic familiarity with plotting and graphing systems in R will be helpful. In the workshop, I will be using the ggplot2 package for visualisation, but a few alternatives exist.

New attendees are always welcome, though I would recommend people take a look at the workshop material provided in the Files section of the meetup group. Feel free to get in contact with me should you have any questions regarding the material in the workshop or the files I have provided.

There will be a small entrance fee to cover the cost of the venue.


Join or login to comment.

  • Mick C.

    The github repository has been updated with some of the example code for the worksheet.

    I have also updated the workshop.zip file in the Files section.

    I did not create code for every single exercise, but the missing parts are just a matter of changing variable names in the execution. As always, let me know if you have any issues trying it.

    September 20, 2013

  • brennan p

    Would there be any chance revisiting this topic or even looking at related topics like market basket analysis

    September 20, 2013

    • Mick C.

      I don't see why not, it's a fascinating and useful topic.

      The main reason for wrapping it up was I did not want people to feel it was pointless coming to it, and there has been a notable dropoff of attendance for the second and third in the series.

      Also, it is good to cover a bunch of different topics.

      Finally, I am working on some sample code to help people with all the exercises, which I hope to be helpful.

      I would definitely be interested in doing another workshop on this topic in the future though, starting where we left off and focusing mainly on the models, running it through JAGS/BUGS and how to analyse and interpret the output.

      September 20, 2013

    • Mick C.

      I am working on sample code for the worksheet and I realised there is quite a bit of stuff towards the end that we still were not able to cover.

      I think it is good to move on to different topics, but I think a standalone workshop dealing with some of the ins and outs of using bayesian analysis and visualisation is worth considering in a few months.

      There are quite a few interesting things to explore still and I could easily fill a workshop with it, especially if I have most of the code prewritten.

      September 20, 2013

  • Mick C.

    Thanks all for coming. I will have details for the exercise code in the next couple of days. I am thinking I will add everything to a repository on github or something, as that should make the job of correcting mistakes easier.

    September 19, 2013

    • Frank K.

      Sorry I couldn't make it Mick. Last minute call to the wall with free ticket!

      September 19, 2013

    • Mick C.

      No problem at all Frank, I'm in the process of organising some suggested code solutions and will let you know once I have put it up.

      September 19, 2013

  • Mick C.

    I created a github repository with the code and files for the workshop for the eager and impatient

    https://github.com/kaybenleroll/bdaworkshop2013

    I will also update the workshop.zip file once I have sample solution to the rest of the worksheet done.

    1 · September 19, 2013

  • Mick C.

    Also, as mentioned, I have put a poll up in the group about ideas for futures workshops, so any and all feedback is welcome.

    September 19, 2013

  • Eoin

    Last minute thing has cropped up so going to have to miss what I'm sure will be another super session :(

    September 18, 2013

  • Jonathan S.

    Just saw this animated version of Metropolis-Hastings sampling of a posterior distribution blogged on Revolution Analytics. Seems a pretty good way of visualizing what's going on in M-H:
    http://blog.revolutionanalytics.com/2013/09/an-animated-peek-into-the-workings-of-bayesian-statistics.html

    1 · September 11, 2013

  • Mick C.

    Just a quick notification that I have added some sample code for Sections 1-4 of the worksheet. This was suggested by some of the attendees to help get the most out of previous workshops.

    Thanks to Eoin, Neill and Martin for their suggestions and editing with this.

    I do not recommend you just blindly execute the code as is, rather, run it line-by-line, try to figure out what each does and why I am doing it.

    As always, any questions about this or any other aspect of the workshop is more than welcome.

    September 3, 2013

13 went

Your organizer's refund policy for Workshop on Bayesian Data Analysis (Part 3)

Refunds are not offered for this Meetup.

Our Sponsors

People in this
Meetup are also in:

Imagine having a community behind you

Get started Learn more
Henry

I decided to start Reno Motorcycle Riders Group because I wanted to be part of a group of people who enjoyed my passion... I was excited and nervous. Our group has grown by leaps and bounds. I never thought it would be this big.

Henry, started Reno Motorcycle Riders

Sign up

Meetup members, Log in

By clicking "Sign up" or "Sign up using Facebook", you confirm that you accept our Terms of Service & Privacy Policy