Skip to contents

Fit Recruitment Model

Source: R/fit-recruitment.R
bb_fit_recruitment.Rd

Fit heirarchical Bayesian recruitment model using Nimble.

Usage

bb_fit_recruitment(
  data,
  adult_female_proportion = 0.65,
  sex_ratio = 0.5,
  min_random_year = 5,
  year_trend = FALSE,
  year_start = 4L,
  nthin = 10,
  niters = 1000,
  priors = NULL,
  allow_missing = FALSE,
  quiet = FALSE
)

Arguments

data

The data.frame.

adult_female_proportion

A number between 0 and 1 of the expected proportion of adults that are female. If NULL, the proportion is estimated from the data (i.e., Cows ~ Binomial(adult_female_proportion, Cows + Bulls)) and a prior of dbeta(65, 35) is used. This prior can be changed via the priors argument.

sex_ratio

A number between 0 and 1 of the proportion of females at birth. This proportion is applied to yearlings.

min_random_year

A whole number of the minimum number of years required to fit year as a random effect (as opposed to a fixed effect).

year_trend

A flag indicating whether to fit a year trend effect. Year trend cannot be fit if there is also a fixed year effect (as opposed to random effect).

year_start

A whole number between 1 and 12 indicating the start of the caribou (i.e., biological) year. By default, April is set as the start of the caribou year.

nthin

A whole number of the thinning rate.

niters

A whole number of the number of iterations per chain after thinning and burn-in.

priors

A named vector of the parameter prior distribution values. Any missing values are assigned their default values in priors_survival() and priors_recruitment(). If NULL, all parameters are assigned their default priors.

allow_missing

A flag indicating whether to allow unobserved years (placeholder rows with NA measurement columns). When TRUE, the year random effects for unobserved years are sampled from the prior distribution rather than being informed by data, giving predictions that reflect the population-level mean with appropriate uncertainty. Requires year to be fit as a random effect (i.e., not supported with fixed year effects). Not supported with Maximum Likelihood models.

quiet

A flag indicating whether to suppress messages and progress bars.

Value

A list of the Nimble model object, data and mcmcr samples.

Details

If the number of years is > min_random_year, a fixed-effects model is fit. Otherwise, a mixed-effects model is fit with random intercept for each year. If year_trend is TRUE and the number of years is > min_random_year, the model will be fit with year as a continuous effect (i.e. trend) and no fixed effect of year. If year_trend is TRUE and the number of years is <= min_random_year, the model will be fit with year as a continuous effect and a random intercept for each year.

The start month of the Caribou year can be adjusted with year_start.

See also

Other model: bb_fit_recruitment_ml(), bb_fit_survival(), bb_fit_survival_ml()

Examples

if (interactive()) {
  fit <- bb_fit_recruitment(bboudata::bbourecruit_a)
}