Reformulating a Gaussian state space model in matrix form, we obtain expressions for the likelihood function and the smoothing vector that are generally more efficient than the standard recursive algorithm. We also retrieve filtering weights and deal with data irregularities.
Efficient matrix approach for classical inference in state space models
Petrella, Ivan
2019-01-01
Abstract
Reformulating a Gaussian state space model in matrix form, we obtain expressions for the likelihood function and the smoothing vector that are generally more efficient than the standard recursive algorithm. We also retrieve filtering weights and deal with data irregularities.
File in questo prodotto:
| File | Dimensione | Formato | |
|---|---|---|---|
|
EL_2019.pdf
Accesso riservato
Dimensione
427.64 kB
Formato
Adobe PDF
|
427.64 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
