ngme.fisher: Estimation of Fisher information matrix

In davidbolin/ngme: Linear Mixed Effects Models With Flexible Distributions

Estimation of Fisher information matrix

Description

Estimates Fisher information matrix for ngme result.

Usage

ngme.fisher(
  fit,
  std.threshold = NULL,
  observed = TRUE,
  silent = FALSE,
  n.cores = 1,
  nSim = 1000,
  n.rep = 10,
  nIter = NULL,
  only.effects = TRUE,
  nBurnin = 100
)

Arguments

fit	An ngme object.
std.threshold	A threshold for the MC standard deviation of the estimates. The estimation is run until all diagonal elements F.i of the inverse Fisher information matrix satisfy std(F.i)*std.threshold < F.i
observed	If TRUE, the observed Fisher information matrix is estimated. Otherwise the ordinary Fisher-Information matrix is estimated.
silent	If TRUE, some diagnostic information is shown during estimation.
n.cores	The number of cores to use for the estimation. n.cores estimates are compued in parallel.
nSim	A numeric value for the number of samples of the Gibbs sampler that is used internally.
n.rep	A numeric value for the numer of MC estimates the initial estimate should be based on. If std.threshold is used further estimates are computed until the criteria are satisfied.
nIter	A numeric value for the number of iterations to be used to obtain the Fisher-Information matrix. If the observed Fisher information matrix is estimate, nIter should be 1 since no new data has to be simulated.
only.effects	If TRUE, the criteria for std.threshold is only applied to the fixed effect estimates.
nBurnin	The number of samples to discard as burnin in the Gibbs sampler.

Value

The ngme result object with the Fisher information added.

Examples

  ## Not run: 
  data(srft_data)
  ngme(...)
  
## End(Not run)
  @export

davidbolin/ngme documentation built on Dec. 5, 2023, 11:48 p.m.