Skip to contents

Plots residuals over time and facets by antigen-isotype (Iso_type). The mean absolute residual for each facet is annotated in the upper-right corner. Fitted and residual values are calculated on demand via calc_fit_mod(), using the original_data, strat, and decay_type attributes stored on model by run_serodynamics(), and include both natural-scale and log10-scale medians and 2.5%/97.5% posterior quantiles.

Usage

plot_residuals(
  model,
  ids = NULL,
  antigen_isos = NULL,
  log_y = TRUE,
  show_interval = TRUE,
  connect_lines = FALSE
)

Arguments

model

An sr_model object (returned by run_serodynamics()), with original_data, strat, and decay_type attributes (see calc_fit_mod()).

ids

(Optional) Participant IDs to include. When supplied, points (and, if connect_lines = TRUE, lines) are colored by subject; otherwise no color is used.

antigen_isos

(Optional) Antigen-isotypes (antigen_iso) to include.

log_y

logical; if TRUE (default), plots the residual computed on the log10 scale (log10(observed) - log10(fitted)); if FALSE, plots the natural-scale residual.

show_interval

logical; if TRUE (default), draws an error bar around each residual spanning its 2.5%/97.5% posterior interval, to visualize the precision of the posterior.

connect_lines

logical; if TRUE, connects each subject's residuals over time with a line. Default FALSE.

Value

A ggplot2::ggplot object.

Examples

plot_residuals(
  model = serodynamics::nepal_sees_jags_output,
  ids = c("sees_npl_128", "sees_npl_131"),
  antigen_isos = c("HlyE_IgA", "HlyE_IgG")
)