Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Move Kalman functionality to ElOceanografo/StateSpace.jl #54

Open
4 tasks
GordStephen opened this issue Oct 21, 2015 · 2 comments
Open
4 tasks

Move Kalman functionality to ElOceanografo/StateSpace.jl #54

GordStephen opened this issue Oct 21, 2015 · 2 comments

Comments

@GordStephen
Copy link
Contributor

That way TimeModels.jl can just focus on specifying particular time series models (ARIMA-family, structural time series, GARCH, etc) and providing algorithms and a nice interface for parameter estimates. More generally this would avoid duplicating development efforts as well.

Elements that would be removed here and rely on StateSpace.jl:

  • StateSpaceModel type
  • Standalone Kalman filtering (faster than standalone Kalman smoothing for non-EM parameter fitting)
  • Kalman smoothing given a filtered model (currently implemented via Rauch-Tung-Striebel. Not a dependency for other parts of the package)
  • Standalone Kalman smoothing (Currently implemented via Durbin-Koopman. Necessary for EM parameter fitting)
@milktrader
Copy link
Contributor

I'm personally in favor of making TimeModels more modest with (as mentioned above) support for ARIMA, GARCH, and perhaps some tests that haven't found a home in StatsBase.

I have a list of those tests somewhere ...

@milktrader
Copy link
Contributor

#23

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants