64th ISI World Statistics Congress - Ottawa, Canada

64th ISI World Statistics Congress - Ottawa, Canada

Post-mortem interval: a functional data analysis for criminology

Author

FF
Frederic Ferraty

Co-author

  • D
    Davide Pigoli
  • J
    John A.D. Aston
  • A
    Anjali Mazumder
  • C
    Cameron Richards
  • M
    Martin J.R. Hall

Conference

64th ISI World Statistics Congress - Ottawa, Canada

Format: IPS Abstract

Keywords: criminology, dynamic-model, functional-data, functional-data-analysis, post-mortem-interval

Session: IPS 428 - Functional data in real life

Tuesday 18 July 10 a.m. - noon (Canada/Eastern)

Abstract

When a body is discovered at a crime scene, it is necessary to determine the time since death or, more formally, the post-mortem interval (PMI). If, in addition, the body has been left outdoors for a long period of time, forensic entomology can be used to estimate this post-mortem interval by examining evidence obtained from the growth of insect larvae on the body. The hatching period of the larvae plays an important role in this methodology as it corresponds to the date of abandonment of the corpse (the latter being assumed to be close to the victim's death). In this presentation, we propose a method for estimating the hatching date of larvae (or maggots) based on their length, the temperature profile of the crime scene and experimental data on larval development. This method requires the estimation of time-dependent growth curve from experiments where the larvae were exposed to a relatively small number of constant temperature profiles. As temperature influences the rate of development, a crucial step is the temporal alignment of the curves at different temperatures. We then propose a dynamic model for the time-varying temperature profiles, based on the local growth rate estimated from the experimental data. This allows us to determine the most likely time of hatching (and hence the PMI) for a sample of larvae from the crime scene of two criminal cases.