Category Archives: EcoGillespie

I currently have code that automatically applies parameter values from a hash map to a model by reflectively looking up the appropriate setter methods. I mentioned that I’d rewrite this to use the reflection API to directly set fields. In … Continue reading →

Today I figured out an interface for having agents (individuals (models)) specify events in EcoGillespie. Each possible event will be, surprise surprise, an object. I considered a few ways of structuring this object. The following were rejected for fairly obvious … Continue reading →

I haven’t even begun writing a single-thread version of the EcoGillespie, but the goal for this project is to insulate the modeler from the details of execution—including, hopefully, leaving open the possibility of multithreaded execution, and ultimately even execution on … Continue reading →

Ha! It turns out I was doing too much work with my automatic parameter setting. Turns out the reflection API has methods for getting and setting field values directly. Very convenient for setting parameters, methinks. I will rework it using … Continue reading →

I did two things with EcoGillespie development today: Renamed the Individual class Node, to reflect that it’s the superclass for anything that can be a node in a network. Wrote a temporary main() function in the DiseaseNetwork example that actually … Continue reading →

The last post was heavy on the theory of the Gillespie algorithm and not so much on the details of the software. So here’s what I’ve come up with so far. Class hierarchy The root object for all objects created … Continue reading →

Last fall, I implemented a model for Mercedes Pascual that, inspired by some models by David Alonso, uses an exact, “event-based” stochastic simulation of disease spreading on a network rather than the usual (and epistemologically unsatisfying) discrete-time approximation. Rather than … Continue reading →