jsdmstan-extractors: Extract quantities useful for model summaries

In fseaton/jsdmstan: Fitting jSDMs in Stan

Extract quantities useful for model summaries

Description

These are methdos for extracting various useful summaries from models, including the model parameter names, NUTS parameters, the log posterior, r-hat and n-eff ratio.

Usage

get_parnames(object, log_lik = FALSE)

## S3 method for class 'jsdmStanFit'
nuts_params(object, ...)

## S3 method for class 'jsdmStanFit'
log_posterior(object, ...)

## S3 method for class 'jsdmStanFit'
rhat(object, ...)

## S3 method for class 'jsdmStanFit'
neff_ratio(object, ...)

Arguments

object	The jsdmStanFit model object
log_lik	Whether the log_lik parameters should be included in the parameter list
...	Arguments passed on to the bayesplot equivalent for stanFit objects

Value

get_parnames() returns a character vector of model parameter names.

nuts_params() returns a molten data frame (see reshape2::melt()). The data frame should have columns "Parameter" (factor), "Iteration" (integer), "Chain" (integer), and "Value" (numeric).

log_posterior() returns a molten data frame (see reshape2::melt()). The data frame should have columns "Chain" (integer), "Iteration" (integer), and "Value" (numeric).

rhat(), neff_ratio() both return named numeric vectors.

Functions

• get_parnames(): Get the model parameter names

• nuts_params(jsdmStanFit): Get the NUTS parameters

• log_posterior(jsdmStanFit): Get the log posterior

• rhat(jsdmStanFit): Get the R-hat

• neff_ratio(jsdmStanFit): Get the n_eff ratio

fseaton/jsdmstan documentation built on Sept. 29, 2024, 6:40 p.m.