est_naloxone {bennu}

Run Bayesian estimation of naloxone number under-reporting

Description

Samples from Bayesian model using input from data frame

Usage

est_naloxone(
  d,
  psi_vec = c(0.7, 0.2, 0.1),
  max_delays = 3,
  delay_alpha = 2,
  delay_beta = 1,
  run_estimation = TRUE,
  rw_type = 1,
  chains = 4,
  iter = 2000,
  seed = 42,
  adapt_delta = 0.85,
  ...
)

Arguments

d

data frame with format

regions

unique id for region

times

time in months

Orders

Kits ordered

Reported_Used

Kits reported as used

Reported_Distributed

Kits reported as distributed

region_name

Optional label for region

psi_vec

reporting delay distribution

max_delays

maximum delay from kit ordered to kit distributed

delay_alpha

shape parameter for order to distributed delay distribution

delay_beta

shape parameter for order to distributed delay distribution

run_estimation

if TRUE will sample from posterior otherwise will sample from prior only

rw_type

1 - random walk of order one. 2 - random walk of order 2.

chains

A positive integer specifying the number of Markov chains. The default is 4.

iter

A positive integer specifying the number of iterations for each chain (including warmup). The default is 2000.

seed

Seed for random number generation

adapt_delta

(double, between 0 and 1, defaults to 0.8)

...

other parameters to pass to rstan::sampling

Value

An S4 rstan::stanfit class object containing the fitted model

See Also

Other inference: est_naloxone_vec()

Examples

## Not run: 
library(rstan)
library(bayesplot)

rstan_options(auto_write = TRUE)
options(mc.cores = parallel::detectCores(logical = FALSE))

d <- generate_model_data()
fit <- est_naloxone(d, iter = 100, chains = 1)
mcmc_pairs(fit,
  pars = c("sigma", "mu0"),
  off_diag_args = list(size = 1, alpha = 0.5)
)

## End(Not run)