integrated_data: create integrated data objects

In jdyen/greta.integrated: Integrated Population Models in greta

Description

integrated_data defines a data object with custom likelihood based on a process model defined with integrated_process

Usage

add_data(data, process, likelihood, bias = no_bias(),
  settings = list(), predictors = NULL)

age_abundance(distribution = poisson)

stage_abundance(distribution = poisson)

age_cjs(distribution = binomial)

stage_cjs(distribution = binomial)

stage_to_age(distribution = multinomial)

age_to_stage(distribution = multinomial)

occupancy(classes, density = no_density(), priors = list())

abundance(distribution = poisson)

community(distribution = bernoulli)

is.integrated_data(object)

## S3 method for class 'integrated_data'
print(x, ...)

## S3 method for class 'integrated_data'
summary(object, ...)

Arguments

data	a single data input (see details for descriptions of possible input types)
process	an integrated_process object
likelihood	something
bias	a bias function that connects observed and modelled data (see details)
settings	a named list of settings passed to data formatting functions (see details)
predictors	object of class integrated_predictor
object	an integrated_data object
x	an integrated_data object
...	additional arguments to print and summary methods (currently ignored)

Details

Do something. The settings list can be used to specify how the data are binned, either with specific breaks for binning or with the number of breaks to use. If these are not provided, the functions use the classes element of process to determine the number of bins (nbreaks = classes + 1).

Value

An object of class integrated_data, which contains information on the data module and can be passed to integrated_model

Examples

## Not run: 

library(integrated)

# prepare an example model
data <- add_data()
                        
# summarise data module
model
summary(model)
plot(model)

## End(Not run)

jdyen/greta.integrated documentation built on May 3, 2019, 8:04 p.m.