Function to estimate a Bayesian case-control model.

Function to estimate a Bayesian case-control model.

Usage

fit(
  data = data_sim,
  contamination = F,
  show_code = T,
  offset = F,
  beta_prior = "normal",
  small_area_prior = "fixed",
  large_area_prior = NA,
  intercept_scale = NA,
  iter = 100,
  warmup = 50,
  thin = 4,
  cores = 4,
  chains = 4,
  control = list(max_treedepth = 7, adapt_delta = 0.8),
  verbose = TRUE
)

Arguments

data

list, survey and shapefile combined list object created with the data.prep() function.

contamination

logical, should the model include the Rota et al. (2013) style contamination layer? Defaults to TRUE.

show_code

logical, should the stan code print on the screen at the end of the run? Defaults to TRUE.

offset

logical, should the model include a King and Zeng (2001) style offset ? if contamination is also specified, this will be a contaminated-offset. Defaults to TRUE.

beta_prior

string, what prior should the regression coefficients have? Choice between: "normal" and "cauchy."

small_area_prior

string, what should be the small-area effects type? Choice between: "fixed," "random," "ICAR," and "BYM2." Specify NA for no area effects. Defaults to "fixed.

large_area_prior

string, what should be the large-area effects type? Choice between: "fixed" and "random." Specify NA for no area effects. Defaults to NA.

intercept_scale

integer, scale of intercept; see ?stan for further information.

iter

integer, number of draws; see ?stan for further information.

warmup

number of warmup draws; see ?stan for further information.

thin

integer, number to thin draws; see ?stan for further information.

cores

integer, number of cores to use; see ?stan for further information.

chains

integer, number of chains; see ?stan for further information.

control

list; tree depth and adapt delta; see ?stan for further information.

verbose

logical, print full stan output? Defaults to TRUE.