Compute table with models result stats.

Summarizes fitted growth models per group, computing R-squared, fit evaluation (good/bad), and completeness at start/end of observed data.

Usage

tbl_models_stats(nc_obj, thresholds_r2 = 0.8, thresholds_asymptote = 0.07)

Arguments

nc_obj

A nowcast_results object.

thresholds_r2

Numeric scalar. R-squared threshold above which (combined with end completeness near 1) a model is classified "Good Fit".

thresholds_asymptote

Numeric double. Gap from 100% Completeness tolerated to end_completeness_pred to categorise the model with good or bad fit.

Value

A tibble with one row per fitted model, including columns for group variables, R2, RSS, model coefficients (a, b, c), completeness at start/end, and eval ("Good Fit"/"Bad Fit"). If nc_obj@models has no rows, it is returned unchanged.

Examples

if (FALSE) { # \dontrun{
tbl_models_stats(nc_obj)
tbl_models_stats(nc_obj, thresholds_r2 = 0.9, thresholds_asymptote = 0.05)
} # }