BibTeX
@MISC{Doucet11generalstate-space,
author = {Arnaud Doucet},
title = {General State-Space Models},
year = {2011}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
State-space models also known as Hidden Markov models are ubiquitous time series models in ecology, econometrics, engineering, statistics etc. Let fXngn1 be a latent/hidden Markov process de ned by X1 µθ () and Xn j (Xn1 = xn1) fθ ( j xn1). We only have access to a process fYngn1 such that, conditional upon fXngn1, the observations are statistically independent and Yn j (Xn = xn) gθ ( j xn). θ is an unknown parameter of prior density p (θ).
