logLik.PF_EM: Approximate Log-Likelihood from a Particle Filter

In dynamichazard: Dynamic Hazard Models using State Space Models

Approximate Log-Likelihood from a Particle Filter

Description

Computes the approximate log-likelihood using the forward filter clouds. See the vignette("Particle_filtering", "dynamichazard") for details.

Usage

## S3 method for class 'PF_EM'
logLik(object, ...)

## S3 method for class 'PF_clouds'
logLik(object, df = NA_real_, nobs = NA_integer_, ...)

Arguments

object	an object of class PF_clouds or PF_EM.
...	unused.
df	degrees of freedom used in the model.
nobs	integer with number of individuals used to estimate the model.

Value

The approximate log-likelihood value given the observed data and set of parameter used when simulating the clouds. An attribute "P(y_t|y_{1:(t-1)})" has the P(y_t|y_{1:(t-1)}) terms.

dynamichazard documentation built on Oct. 6, 2022, 1:08 a.m.