5 thoughts on “Running NetLogo from R: a tutorial

  1. Thanks for the first part, but I seriously doubt the last two sentences of your post. If the parameter sweep would be smooth enough to use “optim” then you probably didn’t need to use ABM in the first place.

    • Well, “optim” has worked well for me: some of the options on “optim” don’t actually require a smooth fitness landscape. It works even better if you can rule out some regions of the parameter space ahead of time. However, given the stochasticity of an ABM, “optim” does require doing several runs at each parameter combination, in order to reduce the noise on the result. That slows things down, although parallelism can really help with that.

  2. Pingback: NetLogo post summary | Scientific Gems

  3. Thanks, Tony, for your tutorial which proves very helpful to me.
    Unfortunately though, I’m not able to fully retrace your commands. Apparently, the Life.nlogo model from the model library has been altered, so that the R commands that refer to the NetLogo model don’t work (e.g. there is no reporter called “history” in the model anymore). Could you please upload the model that you worked with or post a link to it?

    Thanks a lot,

    • Sorry about that. I was not actually using Life.nlogo from the model library, but a simplified version of this model (with, if I recall correctly, “setup-life” named to “setup”, and “average-life-density” renamed to “average-density”).

Leave a Reply to Joanna Bryson Cancel 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

This site uses Akismet to reduce spam. Learn how your comment data is processed.