Plot Model Parameters vs Likelihood (Deprecated)

Deprecated. This parameter-vs-likelihood scatter is redundant with the principled parameter diagnostics already produced by render_MOSAIC_figures: the "sensitivity" group (calc_model_parameter_sensitivity / HSIC importance) and the "posterior" group (prior/posterior densities per parameter). It also scales poorly — a per-facet LOESS over the full retained sample (tens of thousands of simulations across all parameters) makes it the dominant cost of a render pass. It is therefore no longer wired into render_MOSAIC_figures() and will be removed in a future release.

Usage

plot_model_parameters(results, output_dir, verbose = TRUE)

Arguments

results

A data frame containing calibration results with columns for parameters, likelihood values, and simulation identifiers (sim, iter, seed). This is typically the output from a model calibration run. The function will automatically infer the number of simulations and iterations from the data.

output_dir

Character string specifying the directory where plots should be saved. Directory will be created if it doesn't exist.

verbose

Logical indicating whether to print messages. Default is TRUE.

Value

Invisibly returns a list containing the plot objects:

• global: ggplot object for global parameters

• location: Named list of ggplot objects for each location

Details

Creates faceted plots showing the relationship between model parameters and likelihood values from calibration results. Generates separate plots for global and location-specific parameters.

This function creates publication-quality faceted plots showing how each parameter relates to the model likelihood. Points are colored by density, with a LOESS smooth curve showing the trend. The best parameter value (highest likelihood) is highlighted with a red point.

The function automatically:

• Separates global and location-specific parameters

• Categorizes parameters by type (epidemiological, environmental, etc.)

• Creates separate plots for each location found in the data

• Saves plots as PDF files in the specified output directory

Examples

if (FALSE) { # \dontrun{
# Run calibration and collect results
results <- run_MOSAIC()

# Create plots (automatically infers n_sim and n_iter from results)
plots <- plot_model_parameters(results, output_dir = "calibration_output")

# Access individual plots
print(plots$global)
print(plots$location[["ETH"]])
} # }