Model log-likelihood — model_log_likelihood • corehydror

Evaluate a model's log-likelihood, decomposed into its data and prior halves.

Usage

model_log_likelihood(model, params = NULL)

Arguments

model: a corehydro_model from one of the model_*() constructors.
params: optional numeric vector of parameter values; the model's own current values are used when omitted.

Value

A named list: log_likelihood, data_log_likelihood, prior_log_likelihood, and parameters (the vector the model actually evaluated, which a mixture model rewrites in place to normalize its weights).

Examples

peaks <- c(12500, 15300, 8900, 22100, 18700, 14200, 9800, 28500, 17400, 11600)
m <- model_univariate("Normal", peaks)
model_log_likelihood(m)
#> $log_likelihood
#> [1] -135.453
#> 
#> $data_log_likelihood
#> [1] -100.7276
#> 
#> $prior_log_likelihood
#> [1] -34.7254
#> 
#> $parameters
#> [1] 15900.00  6025.87
#> 
model_log_likelihood(m, c(16000, 6000))
#> $log_likelihood
#> [1] -135.4459
#> 
#> $data_log_likelihood
#> [1] -100.7248
#> 
#> $prior_log_likelihood
#> [1] -34.7211
#> 
#> $parameters
#> [1] 16000  6000
#>