EcoDiet (version 2.0.1)

diagnose_model: Diagnose EcoDiet model

Description

This function operates a diagnostic of the fit EcoDiet model.

A message is printed to provide the number of variables for which the Gelman-Rubin diagnostic exceeds specific thresholds (> 1.01, > 1.05, >1.1). The list of the 10 worst variables in terms of convergence is also printed.

You need to have run the run_model function before using this function.

The design of this function is substantially inspired from a function with a similar objective in the MixSIAR package [(Stock et al. 2018)](https://doi.org/10.7717/peerj.5096), for which code is available online on the [MixSIAR GitHub repository](https://github.com/brianstock/MixSIAR). The diagnostic plots are generated using the ggmcmc package [(Fernández-i-Marín, 2016)](https://CRAN.R-project.org/package=ggmcmc).

Usage

diagnose_model(jags_output, var.to.diag = "all", save = FALSE, save_path = ".")

Value

A matrix containing the Gelman diagnostic for all the variables monitored by the run_model function (variables_to_save argument).

Arguments

jags_output

The MCMC output summarized in the class jagsUI object output by run_model function

var.to.diag

The list of variables for which diagnostic plots should be produced and save. By default, this argument is "all" hence is run for all the variables.

save

Indicates whether diagnostic plots should be produced and saved.

save_path

The path indicating where to save the diagnostic plots.

See Also

run_model to run the model

Examples

# \donttest{
realistic_biotracer_data <- read.csv(system.file("extdata", "realistic_biotracer_data.csv",
                                               package = "EcoDiet"))
realistic_stomach_data <- read.csv(system.file("extdata", "realistic_stomach_data.csv",
                                             package = "EcoDiet"))

data <- preprocess_data(biotracer_data = realistic_biotracer_data,
                        trophic_discrimination_factor = c(0.8, 3.4),
                        literature_configuration = FALSE,
                        stomach_data = realistic_stomach_data)
                        
write_model(literature_configuration = FALSE)

mcmc_output <- run_model("EcoDiet_model.txt", data, run_param="test")

Gelman_test <- diagnose_model(mcmc_output)
Gelman_test

# }