In a linear Gaussian state-space time series analysis, a disturbance smoother and a simula-tion smoother are widely used procedures for smoothing and sampling state or disturbance vectors given observations. Several smoothing procedures are also proposed for a non-Gaussian observation process. However, it is assumed that a state equation is linear and that an observation vector and a state vector are conditionally independent. These as-sumptions often need to be relaxed in the analysis of real data. Thus this article considers a general state-space model with a non-Gaussian observation process and a nonlinear state equation where an observation vector and a state vector are allowed to be dependent. We describe a disturbance smoother and a simulation smoother for such models and give numerical examples using simulated data and real data.
Download Info
To our knowledge, this item is not available for
download. To find whether it is available, there are three
options:
1. Check below under "Related research" whether another version of this item is available online.
2. Check on the provider's web page
whether it is in fact available.
3. Perform a search for a similarly titled item that would be
available.
Publisher Info
Paper provided by CIRJE, Faculty of Economics, University of Tokyo in its series CIRJE F-Series with number
CIRJE-F-221.