occ_mod: Fit the single species, single season site occupancy model.

In StrattonCh/occupancy: Fit and Analyze Occupancy Models

Description

This function fits the single species, single season site occupancy model first developed by MacKenzie et al. (2002).

Usage

occ_mod(
  occupancy,
  detection,
  data,
  niter = 1000,
  nchains = 3,
  seed = NULL,
  save_model = FALSE,
  model_name = paste0("occ_model_", Sys.Date()),
  beta_prior = "dnorm(0, 1/2)"
)

Arguments

occupancy	model declaration for the occupancy portion of the model using standard linear model syntax
detection	model declaration for the detection portion of the model using standard linear model syntax
data	data from which to create the detection and covariate matrices. Each row should represent a single visit to a site. See details and examples for proper formatting
niter	number of MCMC iterations
nchains	number of MCMC chains
seed	optional seed
save_model	logical; should a text file containing the model be exported?
model_name	character string defining the name of the text file describing the model if save_model = TRUE
beta_prior	character string defining prior distribution for regression coefficients at the occupancy and detection levels. Priors should be specified using distributions available in NIMBLE. See available distributons in NIMBLE

Details

This function fits the single season, single species site occupancy model using the logit link function. The data should contain columns named site, visit, and y. See examples.

Value

an object of class list containing the following:

• samples object of class list of length nchains, each containing a matrix posterior samples

• loglik object of class list of length codenchains, each containing a matrix of samples of the log posterior likelihood

Examples

# simulate data
sim <- sim_occ(M = 100, max_j = 10, seed = 01012021, rand_visits = FALSE)
data <- sim$data

# note structure of data frame
names(data)
head(data)

# fit model
ex <- occ_mod(occupancy = ~ psi_cov1, detection = ~ p_cov1, data = data,
              niter = 4000, beta_prior = "dunif(-5, -2)")

# results
ex
summary(ex)
str(ex)

StrattonCh/occupancy documentation built on Feb. 17, 2021, 6:36 a.m.