The ultimate goal of the court (and/or jury) is to make a decision concerning a specific suspect’s guilt given the evidence, which in the likelihood ratio paradigm for presenting evidence is usually expressed as the posterior odds in favor of the suspect’s guilt. In this paradigm, the court (and/or jury) is usually responsible for prior beliefs about guilt while the forensic scientist is responsible for providing the likelihood of the evidence when the suspect is guilty (the prosecution proposition) vs. when the suspect is not guilty (the defense proposition).

In this presentation, we will discuss two techniques for estimating the forensic likelihood ratio for handwriting evidence. The first, a parametric approach, assumes that a handwriting profile can be modeled by a high dimensional multinomial distribution. The second, a nonparametric method, makes use of a biometric score. We then compare and contrast various sets of prosecution and defense propositions (and the resulting score-based likelihood ratios) which have appeared in the literature. Our goal in performing these comparisons is to illustrate the effect that subtle modifications of these propositions can have on score-based likelihood ratios.