bayesvl stan utilities {bayesvl}

Build RStan models from directed acyclic graph

Description

Build Stan models from directed acyclic graph of an object of class bayesvl.

Usage

# build Stan models from directed acyclic graph.
bvl_model2Stan(dag, ppc = "")

# compile and simulate samples from the model.
bvl_modelFit(dag, data, warmup = 1000, iter = 5000, chains = 2, ppc = "", ...)

# summarize the stan priors used for the model.
bvl_stanPriors(dag)

# summarize the stan parameters used for the model.
bvl_stanParams(dag)

# summarize the generated formula at the node.
bvl_formula(dag, nodeName, outcome = T, re = F)

Arguments

dag

an object of class bayesvl

data

a data frame or list containing the data

warmup

Optional: Number of warmup iterations. By default, half of iter

iter

Optional: Number of iterations of sampling. Default is 5000

chains

Optional: Number of independent chains to sample from. Default is 2

ppc

Optional: a character string contains posterior predictive check scripts

...

extra arguments from the generic method

nodeName

A character string contains the node name

outcome

Optional: Whether show out distribution

re

Optional: Whether run recursive for all up-level nodes

Value

bvl_model2Stan() return character string of rstan code generated from the model.

bvl_modelFit() return an object class bayesvl which contains result with the following slots.

model

Stan model code

stanfit

stanfit object returned by stan

standata

The data

pars

Parameter names monitored in samples

formula

Generated formula from the model

bvl_stanPriors() return character string of rstan priors generated from the model.

bvl_stanParams() return character string of rstan parameters generated from the model.

Author(s)

La Viet-Phuong, Vuong Quan-Hoang

References

For documentation, case studies and worked examples, and other tutorial information visit the References section on our Github:

• github.com/sshpa/bayesvl/master/References/bvl_ug_en08.pdf

Examples

# Design the model in directed acyclic graph
model <- bayesvl()
model <- bvl_addNode(model, "Lie", "binom")
model <- bvl_addNode(model, "B", "binom")
model <- bvl_addNode(model, "C", "binom")
model <- bvl_addNode(model, "T", "binom")

model <- bvl_addArc(model, "B", "Lie", "slope")
model <- bvl_addArc(model, "C", "Lie", "slope")
model <- bvl_addArc(model, "T", "Lie", "slope")

# Generate the Stan model's code
model_string <- bvl_model2Stan(model)
cat(model_string)

# Show priors in generated Stan model
bvl_stanPriors(model)