lir_model_selection: Model selection for models of lagged identification rate

In Alexhaoge/rCLIFII: Composite Likelihood Inference for Individual Identifications

Model selection for models of lagged identification rate

Description

Model selection for models of lagged identification rate

Usage

lir_model_selection(
  X,
  n,
  tp,
  model,
  method,
  ncores = 4,
  mtau = 1000,
  nboot = -1,
  bin_len = -1,
  model_cl_fun = NULL,
  cl.H = NULL,
  model.K = NULL,
  seed = NULL
)

Arguments

X	A list or matrix containing the identities of individuals identified in each sampling period
n	A vector or a positive integer, representing the number of individuals identified in each sampling period It indicates the same number of individuals identified in all sampling periods if a positive integer
tp	A set of observed time
model	Models of lagged identification rate, model = 'lir_1', 'lir_2', 'lir_3', or formulate model by yourself 'model_cl_fun'
method	The method = 'Bootstrap', 'BBootstrap', or 'Jackknife'
ncores	doParallel.
mtau	The maximum allowable lag time.
nboot	The number of bootstrap samples desired
bin_len	An integer represents len-time-unit intervals
model_cl_fun	If you formulate your model, please input function to calculate the composite likelihood about your model
cl.H	If you formulate your model, please input the sensitivity matrix with respect to parameters in your model
model.K	If you formulate your model, please input the number of parameters in your model
seed	Random seed.

Details

See Akaike (1973) for Akaike information criterion (AIC); See Burnhan and Anderson (2002) for Quasi-Akaike information criterion (QAIC); See this paper for composite likelihood information criterion (CLIC).

Value

The values of model selection criteria.

Alexhaoge/rCLIFII documentation built on Sept. 28, 2023, 11:23 p.m.