bea_count: Bayesian exposure assessment from count data

In brechtdv/QMRA: Parametric Models for Quantitative Microbial Risk Assessment

Description

Bayesian parametric modelling of exposure concentration from count data.

Usage

bea_count(x, q = 1, data,
          model = c("poisson", "p",
                    "negbin", "nb",
                    "poislognorm", "pln",
                    "poisweibull", "pw"),
          inits = NULL, nchains = 2, burnin = 5000, update = 5000,
          verbose = FALSE)

Arguments

x	The observed counts.
q	The quantities (e.g., volumes, masses) in which the counts were observed; defaults to 1.
data	An optional data frame, containing the variables in the model. If not found in data, the variables are taken from the environment from which ea_count is called.
model	A character string naming the count model to be fitted. See details below.
inits	Named list of initial values; defaults to NULL, leading JAGS to generate inits automatically (see jags.model).
nchains	Number of model chains, should be ≥ 2; defaults to 2.
burnin	Number of samples to discard as burn-in; defaults to 5000.
update	Number of samples to retain; defaults to 5000.
verbose	Should JAGS process info be printed to the R console? defaults to FALSE.

Details

Distributions available:

• Poisson: model = "poisson" or "p"

• Negative Binomial: model = "negbin" or "nb"

• Poisson-LogNormal: model = "poislognorm" or "pln"

• Poisson-Weibull: model = "poisweibull" or "pw"

Value

An object of class "bea".

Author(s)

[e-mail]

References

• Haas CN, Rose JB, Gerba CP (1999) Quantitative Microbial Risk Assessment. John Wiley & Sons, Inc.

See Also

bea_presence, for modelling exposure from presence/absence data
bea_conc, for modelling exposure from concentration data

Examples

## Bayesian exposure assessment from count data
bea_count(x = counts, q = volume, model = "negbin", data = crypto,
          inits = list(shape = 1, rate = 1))

brechtdv/QMRA documentation built on May 13, 2019, 5:06 a.m.