Approximate forward–backward algorithm for a switching linear Gaussian model

A hidden Markov model with two hidden layers is considered. The bottom layer is a Markov chain and given this the variables in the second hidden layer are assumed conditionally independent and Gaussian distributed. The observation process is Gaussian with mean values that are linear functions of the...

Celý popis

Uložené v:

Podrobná bibliografia
Vydané v:	Computational statistics & data analysis Ročník 55; číslo 1; s. 154 - 167
Hlavní autori:	Hammer, Hugo, Tjelmeland, Håkon
Médium:	Journal Article
Jazyk:	English
Vydavateľské údaje:	Amsterdam Elsevier B.V 2011 Elsevier
Edícia:	Computational Statistics & Data Analysis
Predmet:	Algorithms Approximation Approximation Forward-backward algorithm Hidden Markov model Metropolis-Hastings algorithm Seismic inversion Computation Computer simulation Exact sciences and technology Forward–backward algorithm Gaussian General topics Hidden Markov model Markov processes Mathematical analysis Mathematical models Mathematics Metropolis–Hastings algorithm Multivariate analysis Numerical analysis Numerical analysis. Scientific computation Numerical methods in probability and statistics Probability and statistics Probability theory and stochastic processes Sciences and techniques of general use Seismic inversion Statistics Switching Hidden Markov model Approximation Metropolis–Hastings algorithm Seismic inversion Forward–backward algorithm Density estimation Conditional distribution Mixing Statistical distribution Iterative method Dynamical system Metropolis-Hastings algorithm Stochastic method Linear model Markov chain Direct method Distribution function Convergence rate Approximation theory Mixed distribution Data analysis Linear function Convergence acceleration Numerical linear algebra Metropolis Hastings algorithm Mean value Forward-backward algorithm Value function Numerical analysis Gaussian process Linear system Statistical computation Simulation Matrix inversion Hidden Markov models Gaussian processes
ISSN:	0167-9473, 1872-7352
On-line prístup:	Získať plný text
Tagy:	Pridať tag Žiadne tagy, Buďte prvý, kto otaguje tento záznam!

Buďte prvý, kto okomentuje tento záznam!