Using correlated binomial distribution in estimating error rates for forensic firearm identification

• Activity Number: 511 - Statistical Applications in the Physical Sciences
• Type: Contributed
• Date/Time: Wednesday, July 31, 2019 : 10:30 AM to 12:20 PM
• Sponsor: Section on Physical and Engineering Sciences
• Author(s): Nien-Fan Zhang*
• Companies: NIST
• Keywords: Ballistic signatures; Bernoulli trials; forensic science; maximum likelihood estimation; nonlinear regression

Abstract

Estimating error rates for firearm evidence identification is a fundamental challenge in forensic science. The recently developed Congruent Matching Cells (CMC) method provides applications to firearm evidence identification. To estimate error rates, appropriate statistical models are needed for the CMC values. In this paper, in addition to the binomial probability distribution the correlated binomial distribution is proposed. For an image comparison, correlated binomial distribution can be applied to the cell pairs from CMC method. An application to an actual data set demonstrates that the correlated binomial distribution fits the relative frequency distribution of CMC values much better than the binomial distribution.