Roadmap for Julia push

[albop]

@anasza, @jagman88, @spencerlyon2

I have just merged into master, our work with @anasza. There is one implementation of time_iteration_direct, time_iteration, value_iteration which all work for any type of shock, thanks to new decision rules objects (look in test/test_algos.jl ) . Modulo a few possible improvements, it feels just like the right approach. I have listed below the pieces we still need to implement to get a fully working version: (more details in https://github.com/EconForge/Dolo.jl/wiki/ROADMAP)

• [ ] Simulate methods for all processes
  • [x] IID
  • [x] AR1
  • [x] Markov chains
• [ ] Simulate a model with a solution
  • [x] IID
  • [x] AR1
  • [x] Markov chains
• [x] AR1 decision rule
• Solution algorithms
  • [x] VFI (cleanup)
  • [ ] PEA
  • [x] perturbation (requires fix to AR1)
• [ ] Distribution object
• Ergodic distribution
  • [ ] Simulation based
  • [ ] By finding the fixed point of a discrete Markov Chain.
• [ ] Accuracy measures

[sglyon]

This is really cool! I don't think the merged master branch has been pushed to github yet, but I checked it out on the other branch.

I don't have the time to go through it carefully right now, but I will be doing that at some point.

[albop]

Indeed, I didn't push the merged version. Done.