plot_decision_curve: Plot Decision Curve (FPR vs FNR)

In mispitools: Missing Person Identification Tools

Plot Decision Curve (FPR vs FNR)

Description

Creates a scatter plot showing the trade-off between false positive rate (FPR) and false negative rate (FNR) across different LR threshold values. This visualization helps identify optimal decision thresholds based on the relative costs of different types of errors.

Usage

plot_decision_curve(datasim, LRmax = 1000)

Arguments

datasim	A data.frame with columns Related and Unrelated containing LR values. Can be output from sim_lr_genetic, sim_lr_prelim, lr_to_dataframe, or lr_combine.
LRmax	Numeric. Maximum LR value to use as threshold. Points are generated for thresholds from 1 to LRmax. Default: 1000.

Details

If the input is a list (output from sim_lr_genetic), it is automatically converted to a data.frame using lr_to_dataframe.

Error Rate Definitions:

• FPR: Proportion of unrelated (H2) cases with LR > threshold

• FNR: Proportion of related (H1) cases with LR < threshold

Ideal point: The origin (0,0) represents perfect discrimination. Points closer to the origin indicate better thresholds.

Trade-off: Moving along the curve, decreasing FNR typically increases FPR and vice versa. The optimal point depends on the relative costs of false positives vs false negatives.

Value

A ggplot2 scatter plot where:

• X-axis: False Negative Rate (FNR) - proportion of true matches missed

• Y-axis: False Positive Rate (FPR) - proportion of non-matches incorrectly identified

• Each point represents a different LR threshold

The first and last threshold values are labeled on the plot.

References

Marsico FL, Vigeland MD, Egeland T, Herrera Pinero F (2021). "Making decisions in missing person identification cases with low statistical power." Forensic Science International: Genetics, 52, 102519. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1016/j.fsigen.2021.102519")}

See Also

plot_lr_distribution for LR distribution visualization, decision_threshold for computing optimal threshold, threshold_rates for error rates at a specific threshold.

Examples

# Using preliminary data
lr_sims <- sim_lr_prelim("sex", numsims = 500, seed = 123)
plot_decision_curve(lr_sims)

# With lower maximum threshold for finer resolution
plot_decision_curve(lr_sims, LRmax = 100)

mispitools documentation built on Aug. 26, 2026, 1:08 a.m.