lr_hair_color: Likelihood Ratio for Hair Color

In mispitools: Missing Person Identification Tools

Likelihood Ratio for Hair Color

Description

Simulates hair color observations and optionally computes likelihood ratios (LRs) under either H1 (unidentified person is the missing person) or H2 (unidentified person is not the missing person).

Hair color is categorized into 5 groups: 1=Black, 2=Brown, 3=Blonde, 4=Red, 5=Gray/White

Usage

lr_hair_color(
  MPc = 1,
  epc = error_matrix_hair(),
  erRc = epc,
  numsims = 1000,
  Pc = c(0.3, 0.2, 0.25, 0.15, 0.1),
  H = 1,
  Qprop = MPc,
  LR = FALSE,
  seed = 1234,
  nsims = NULL
)

Arguments

MPc	Integer (1-5). Missing person's hair color category. Default: 1.
epc	Matrix. Hair color error/confusion matrix, typically created with error_matrix_hair. Rows represent true colors, columns represent observed colors. Default: error_matrix_hair().
erRc	Matrix. Error matrix for the reference/database. Defaults to epc.
numsims	Integer. Number of simulations to perform. Default: 1000.
Pc	Numeric vector of length 5. Hair color proportions in the population. Must sum to 1. Default: c(0.3, 0.2, 0.25, 0.15, 0.1).
H	Integer (1 or 2). Hypothesis to simulate under: 1: H1 (Related) - Unidentified person IS the missing person 2: H2 (Unrelated) - Unidentified person is NOT the missing person Default: 1.
Qprop	Integer. Query color for testing. Defaults to MPc.
LR	Logical. If TRUE, compute and return LR values. Default: FALSE.
seed	Integer. Random seed for reproducibility. Default: 1234.
nsims	Deprecated. Use numsims instead.

Details

Under H1 (Related): Observed color is sampled using the row of the error matrix corresponding to the MP's true hair color. This accounts for observation errors.

Under H2 (Unrelated): Color is sampled from the population proportions Pc.

LR Calculation: LR = P(observed color | true color is MPc) / P(observed color in population) LR = epc(MPc, observed) / Pc(observed)

Value

A data.frame with column Col containing simulated color observations (1-5). If LR = TRUE, also includes column LRc with the likelihood ratio for each observation.

Deprecation

Soft-deprecated in mispitools 2.0. This feature is generalised by nongenetic_feature (a categorical feature with a full confusion matrix); the unified per-feature engine reproduces epc[MPc, o] / Pc[o] exactly. The legacy function still works for the 2.0 release-candidate cycle and will be removed afterwards.

References

Marsico FL, et al. (2023). "Likelihood ratios for non-genetic evidence in missing person cases." Forensic Science International: Genetics, 66, 102891. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1016/j.fsigen.2023.102891")}

See Also

nongenetic_feature for the unified replacement, error_matrix_hair for creating the error matrix, lr_pigmentation for combined pigmentation traits, sim_lr_prelim for unified preliminary LR simulations.

Examples

# Simulate under H1 (related) with black hair MP
sim_h1 <- lr_hair_color(MPc = 1, H = 1, numsims = 100)
table(sim_h1$Col)

# Simulate under H2 with LR values
sim_h2 <- lr_hair_color(MPc = 2, H = 2, numsims = 100, LR = TRUE)
head(sim_h2)
summary(sim_h2$LRc)

# Custom population proportions
sim_custom <- lr_hair_color(
  MPc = 3,  # Blonde
  Pc = c(0.1, 0.4, 0.3, 0.1, 0.1),  # Different population
  numsims = 500,
  LR = TRUE
)

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