The likelihood ratios calculated using the novel PCA approach were compared with previously reported LRs calculated using a more computationally intensive MultivariateKernel (MVK) model, followed by a calibration step using a Pool Adjacent Violators (PAV) algorithm.
Dimensionality reduction of multivariate elemental concentrations of glass is reported for computing likelihood ratios (LRs). The LRs calculated using principal component analysis (PCA) and a post hoc calibration steps result in low (<1 percent) false inclusions when comparing glass samples known to originate from different sources and low (<1 percent) false exclusions when comparing glass samples known to originate from the same source. The current study found that in both cases, the calibrated LRs limited the magnitude of the misleading evidence, providing only weak to moderate support for the incorrect hypotheses. Most of the different pairs that were found to be falsely included were explained by chemical relatedness (same manufacturer of the glass sources in very close time interval between manufacture). The computation of LRs using dimensionality reduction of elemental concentrations using PCA may transfer to other multivariate data-generating evidence types. (publisher abstract modified)
Downloads
Similar Publications
- Neighborhood Disadvantage, Neighborhood Instability, and Adolescent Behavior: Premarital Childbearing, Dropping Out of School, and Delinquency
- The Minnesota Sexual Assault Kit Initiative (SAKI): Lessons Learned from a Decade of SAKI Evaluations
- Lead by Example: The Effects of Police Supervisors on Officer Behavior