Discovery of transition probabilities between

[GoogleCodeExporter]

We are interested in discovering fixed points/attractors in the time series 
results, the first stage of this is the density graphs, second is finding where 
the time series cluster, and third is finding the transition probabilities 
between these clusters.

Implement a method for discovering the transition probabilities overtime. Ie. 
the probability of a agent in state Y being in state X n generations in the 
future. The method should take a set of states States and a window length n as 
inputs and output a transition probability matrix State*States -> boolean.

Original issue reported on code.google.com by [email protected] on 4 Apr 2011 at 8:48