[R] Fwd: The StructTS method

[Prof Brian Ripley]

On 23/03/2012 18:43, Fretheim, Alexander H wrote:
>
>
> To whomever it may concern,
>
> I'm a young Industrial Engineer working on Senior Design at Georgia
> Tech and have found the StructTS method to be excellent for the
> training set for my forecasting project. There's only one problem: I
> don't actually understand what a Structural Time Series IS. I've
> looked up resources on it, and get that essentially you're dividing
> the Time Series in to additive components dependent on time, but have
> no idea how your method works or why. I've also looked at the
> documentation, which is great from a programmers standpoint and gives
> at least a basic format but without any idea as to how optimization
> occurs. Would you be able to even just supply some C code (which I
> could open in notebook, hopefully) showing how the method
> works/optimizes/heuristically recommends and why?

This is what references are for.  From ?StructTS:

References:

      Brockwell, P. J. & Davis, R. A. (1996).  _Introduction to Time
      Series and Forecasting_.  Springer, New York.  Sections 8.2 and
      8.5.

      Durbin, J. and Koopman, S. J. (2001) _Time Series Analysis by
      State Space Methods._ Oxford University Press.

      Harvey, A. C. (1989) _Forecasting, Structural Time Series Models
      and the Kalman Filter_.  Cambridge University Press.

      Harvey, A. C. (1993) _Time Series Models_.  2nd Edition, Harvester
      Wheatsheaf.

The last is the best place to start.  R is *not* a statistics tutorial.

Beyond that, R is Open Source and you can read all the source code for 
yourself.


> Sincerely,
>
> Alexander Fretheim


-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595