8 thoughts on “The EasyABC package for Approximate Bayesian Computation in R

  1. Hello,

    I am trying to implement an ABC MCMC scheme in R similar to your example above. I notice when I try to run your code above, I receive the following errors:

    1) Prior must be in the form of a list.
    2) Tab Normalization must have the same dimensions as the observed summary statistics.

    Do you find the same errors?



    • Hi Henry,

      yes, thanks for letting me know – when I wrote this post the package was brand new and I think Franck and the others must have changed the syntax since then. It’s more flexible now with the priors, I think when I wrote the post only uniform priors were possible.

      I have corrected the code, the example should run again!



  2. Pingback: Explaining the ABC-Rejection Algorithm in R | theoretical ecology

  3. Hi Florian,

    Do you know if the EasyABC package can be used for inference in Dynamic Bayes Nets?

    Thanks for these blogs. They’re very useful.


    • Hi Paul,

      as long as you can simulate from your model, you can always use ABC. The question is if that’s efficient. In a hierarchical dependency structure, I would think that some kind of iterative filtering through the network is more efficient (assuming you have data on intermediate nodes). But it depends on your problem and data, hard to say in general.


      • Great, and thanks, that makes sense. If your dependencies are governed by unknown and non-linear processes (i.e. they come from a generative simulation but you don’t know the corresponding likelihood functions), then you’d be stuck with some type of ABC method, right? Or do you know of a potentially more efficient technique in that circumstance?

        Thanks again.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s