bm2_kalman_logLik: Exact log-likelihood for Brownian motion spatPomp generator...

In kidusasfaw/spatPomp: Inference for Spatiotemporal Partially Observed Markov Processes

Exact log-likelihood for Brownian motion spatPomp generator with shared or unit-specific parameters

Description

Computes the exact likelihood for a model constructed using bm2, using the Kalman filter. This model is useful for testing methods for models with unit-specific parameters, or method such as ibpf which require a unit-specific extension of shared parameters.

Usage

bm2_kalman_logLik(bm2_object, params = coef(bm2_object))

Arguments

bm2_object	A spatPomp model built using bm2.
params	A parameter vector at which to evaluate the log-likelihood. whereas observations occur every one time unit

Value

A numeric value for the log-likelihood.

Author(s)

Edward L. Ionides

Examples

# Further examples are provided in the tests directory
## Not run: 
b <- bm2()
bm2_kalman_logLik(b)

## End(Not run)

kidusasfaw/spatPomp documentation built on Nov. 20, 2024, 9:36 a.m.