Compare Different Value Function Types

Creates a comparison plot showing multiple value function types (Linear, Piecewise Linear, Exponential, Sigmoid, Step) overlaid on the same plot. This visualization helps stakeholders understand how different functional forms would transform the same raw clinical data, and demonstrates why linear functions are the regulatory-preferred default. Creates separate plots for benefits (increasing direction) and risks (decreasing direction).

Usage

compare_value_function_types(
  benefit_name = "Benefits",
  benefit_min = 0,
  benefit_max = 100,
  benefit_label = NULL,
  risk_name = "Risks",
  risk_min = 0,
  risk_max = 50,
  risk_label = NULL,
  n_points = 100,
  show_titles = TRUE,
  show_legend = TRUE,
  power = 2,
  base_font_size = 9
)

Arguments

benefit_name

Character string for the benefit criterion name. Default is "Benefits".

benefit_min

Numeric value for benefit minimum threshold. Default is 0.

benefit_max

Numeric value for benefit maximum threshold. Default is 100.

benefit_label

Character string for benefit x-axis label. If NULL, uses benefit_name. Default is NULL.

risk_name

Character string for the risk criterion name. Default is "Risks".

risk_min

Numeric value for risk minimum threshold. Default is 0.

risk_max

Numeric value for risk maximum threshold. Default is 50.

risk_label

Character string for risk x-axis label. If NULL, uses risk_name. Default is NULL.

n_points

Integer specifying number of points for plotting. Default is 100.

show_titles

Logical indicating whether to show plot titles. Default is TRUE.

show_legend

Logical indicating whether to show the legend. Default is TRUE.

power

Numeric value for the power/exponential function exponent. Default is 2 (risk-averse, diminishing returns).

base_font_size

Numeric; base font size in points for all text elements in the plot (default: 9).

Value

A combined plot (using patchwork) showing value function type comparisons for both benefits and risks side by side.

Examples

# Default comparison
comparison_plot <- compare_value_function_types()

# Custom comparison with specific criteria
custom_comparison <- compare_value_function_types(
  benefit_name = "Efficacy",
  benefit_min = 0,
  benefit_max = 100,
  benefit_label = "Response Rate (%)",
  risk_name = "Safety",
  risk_min = 0,
  risk_max = 50,
  risk_label = "Adverse Event Rate (%)"
)

# Without titles for cleaner display
if (FALSE) { # \dontrun{
comparison_clean <- compare_value_function_types(
  show_titles = FALSE
)
} # }