ASSESSING SPATIAL HETEROGENEITY IN THE REFRACTIVE INDEX OF FLOAT GLASS

The intrinsic variability of the refractive index within a pane of glass is important in the statistical interpretation of forensic glass evidence. Spatial structure could drastically alter the mode of analysis or severely restrict the generalization in interpretation. This article proposes two meth...

Full description

Saved in:
Bibliographic Details
Published in:Jurimetrics (Chicago, Ill.) Ill.), 2005-10, Vol.46 (1), p.31-51
Main Authors: Maimon, Geva, Steele, Russell J., Curran, James M.
Format: Article
Language:English
Subjects:
Bayesian analysis
Datasets
Disease models
Forensic evidence
Forensic sciences
Glass
Hypothesis testing
Light refraction
Monte Carlo simulation
Null hypothesis
P values
Sampling distributions
Spatial models
Statistical variance
SYMPOSIUM: FORENSIC STATISTICS PART I
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The intrinsic variability of the refractive index within a pane of glass is important in the statistical interpretation of forensic glass evidence. Spatial structure could drastically alter the mode of analysis or severely restrict the generalization in interpretation. This article proposes two methods for determining the amount of spatially structured variability of the refractive index in a pane of float glass. The first method is a standard Bayesian spatial analysis that allows one to infer whether there is evidence of spatially structured variability within a pane. Unfortunately, no formal notion of statistical significance can be attached to this standard analysis. Second, we construct a Monte Carlo hypothesis test to test the null hypothesis that there is no spatial structure of the refractive index within the pane of glass. In addition, we present a simulation study that shows one proposed test quantity to be a good gauge of spatial structure when used in conjunction with a Monte Carlo hypothesis test.
ISSN:0897-1277
2154-4344