mmcif_logLik {mmcif}

Evaluates the Log Composite Likelihood and its Gradient

Description

Evaluates the log composite likelihood and its gradient using adaptive Gauss-Hermite quadrature.

Usage

mmcif_logLik(
  object,
  par,
  ghq_data = object$ghq_data,
  n_threads = 1L,
  is_log_chol = FALSE
)

mmcif_logLik_grad(
  object,
  par,
  ghq_data = object$ghq_data,
  n_threads = 1L,
  is_log_chol = FALSE
)

Arguments

object

an object from mmcif_data.

par

numeric vector with parameters. This is either using a log Cholesky decomposition for the covariance matrix or the covariance matrix.

ghq_data

the Gauss-Hermite quadrature nodes and weights to use. It should be a list with two elements called "node" and "weight". A default is provided if NULL is passed.

n_threads

the number of threads to use.

is_log_chol

logical for whether a log Cholesky decomposition is used for the covariance matrix or the full covariance matrix.

Value

A numeric vector with either the log composite likelihood or the gradient of it.

Examples

if(require(mets)){
  # prepare the data
  data(prt)

  # truncate the time
  max_time <- 90
  prt <- within(prt, {
    status[time >= max_time] <- 0
    time <- pmin(time, max_time)
  })

  # select the DZ twins and re-code the status
  prt_use <- subset(prt, zyg == "DZ") |>
    transform(status = ifelse(status == 0, 3L, status))

  # randomly sub-sample
  set.seed(1)
  prt_use <- subset(
    prt_use, id %in% sample(unique(id), length(unique(id)) %/% 10L))

  n_threads <- 2L
  mmcif_obj <- mmcif_data(
    ~ country - 1, prt_use, status, time, id, max_time,
    2L, strata = country)

  # get the staring values
  start_vals <- mmcif_start_values(mmcif_obj, n_threads = n_threads)

  # compute the log composite likelihood and the gradient at the starting
  # values
  mmcif_logLik(
    mmcif_obj, start_vals$upper, is_log_chol = TRUE, n_threads = n_threads) |>
    print()
  mmcif_logLik_grad(
    mmcif_obj, start_vals$upper, is_log_chol = TRUE, n_threads = n_threads) |>
    print()
}