simhet: Conducts simulation of abundance estimation for Huggins or...

In jlaake/DoubleObs: Analysis of double-observer survey data

Description

Generates simulation replicates of double observer survey data with a known (marked) and unmarked portion of the population with or without group sizes. Fits a sequence of models, and reports average abundance, average std error, confidence interval coverage for each model and model average estimate.

Usage

simhet(
  reps,
  n,
  nradios,
  beta,
  gen_formula = "~-1+loc+cov",
  p0 = 0,
  data = NULL,
  fit_formula = "",
  use_gs = FALSE,
  gs_param = c(1, 15),
  which = c(TRUE, TRUE, rep(FALSE, 4)),
  seed = NULL
)

Arguments

reps	number of simulation replicates
n	number of unmarked groups
nradios	number of marked groups
beta	vector of parameters for detection model specified in gen_formula
gen_formula	formula specified as character for generating detection probabilities for simulated data
p0	proportion of population that has 0 probability of detection
data	dataframe containing covariate data if something other than loc or cov (unmodelled heterogeneity); number of rows should match n+nradios
fit_formula	character portion of the formula used to fit to observed double observer data excluding heterogeneity portion; use lngs for log group size
use_gs	if TRUE, generate random group sizes from gamma distribution (integerized) with parameters in gs_param below
gs_param	vector of 2 parameters for gamma distribution; gs_param[1]=scale and gs_param[2]=shape;see ?rgamma
which	logical vector indicating which of the 6 models are fitted to the data
seed	if NULL uses random seed, but if a value is used it can be used to repeat results from same seed

Value

list containing: 1) abundance estimate (either groups or animals) , 2) std error, 3) lower 95 percent confidence limit, 4) upper 95 percent confidence limit, 5) cover = 1 if TrueN in interval, 6) missed_low=1 if TrueN less than lower limit, and 7) missed_high=1 if TrueN greater than upper limit.

Author(s)

Jeff Laake

Examples

# runs 3 replicates as an example with parameters used to generate example_data and
# fits models with no terms other than loc and residual heterogeneity and others
# that added some combination of veg and lngs.

simres=simhet(3,150,25,beta=c(-.5,-.5,-1,2,.5),data=data.frame(veg=runif(175)),
gen_formula="~-1+loc+veg+cov+lngs",fit_formula=c("veg+lngs","lngs","veg",""),
use_gs=TRUE,gs_param=c(1,10),seed=9234397)

jlaake/DoubleObs documentation built on Dec. 21, 2021, 12:12 a.m.