mispitools: Likelihood Ratios in Forensic Sciences

The Problem

When unidentified human remains are found, forensic scientists must search databases of missing persons to find potential identifications. mispitools provides a statistical framework based on likelihood ratios (LRs) to quantify the weight of evidence, combining genetic and non-genetic information, and support decision-making in these investigations.

Video Tutorial

Tutorial by Suisei Nakagawa

Why Likelihood Ratios?

The likelihood ratio is the gold standard for evidence evaluation in forensic science. Rather than providing a binary "yes/no" answer, the LR tells us how much the evidence should change our belief about an identification.

We evaluate evidence under two competing hypotheses:

• H1: The person of interest (POI) is the missing person (MP)
• H2: The POI is not the MP (is an unrelated individual from the population)