Forensic Glass Transfer Probabilities

Description

Statistical interpretation of forensic glass transfer (Simulation of the probability distribution of recovered glass fragments).

Details

Package: tfer
Type: Package
Version: 1.1
Date: 2010-11-07
License: GPL-2
LazyLoad: yes
Depends: methods

The tfer package provides functions for simulating the number of recovered glass fragments given the conditions set by the user on factors affecting the transfer, persistence and recovery of glass fragments. A large simulation size will provide precise estimates of transfer probabilities to be used in the Bayesian interpretation of forensic glass evidence.

The transfer constructor function creates an object of class transfer consisting of a list of simulated number of recovered glass fragments and the input parameters set by the user. This function is based on the full graphical model in Curran et al. (1998). The user can specify arguments for simulation size, distance, transfer, persistence and recovery properties.

The values function extracts the simulated random variates from a transfer object. para returns the input parameters and user-specified arguments as a numeric vector. parameters is an alternative way of displaying the input parameters and arguments. The initial information specified by the user are concatenated and displayed as a string. Users may find this more informative than para as it displays what each parameter denotes.

tprob returns the transfer probabilities for each unique value of the simulated random variates. If the user is only interested in the probabilities of recovering a certain number of fragments, this can be specified as the second argument of tprob.

summary provides summary statistics of transfer objects and returns a list of input parameters, five-number summary and probabilities of transfer.

The user has three plotting options for producing a graphical view of a transfer object. The plot type can be specified as (0 = barplot of probabilities, 1 = barplot of frequencies or 2 = histogram). A barplot of probabilities is set as the default.

Author(s)

James Curran and TingYu Huang

Maintainer: TingYu Huang <thua041@aucklanduni.ac.nz>

References

Curran, J. M., Hicks, T. N. & Buckleton, J. S. (2000). Forensic interpretation of glass evidence. Boca Raton, FL: CRC Press.

Curran, J. M., Triggs, C. M., Buckleton, J. S., Walsh, K. A. J. & Hicks, T. N. (January, 1998). Assessing transfer probabilities in a Bayesian interpretation of forensic glass evidence. Science & Justice, 38(1), 15-21.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
library(tfer)

## create a transfer object using default arguments
y = transfer()

## probability table
probs = tprob(y)

## extract the probabilities of recovering 8 to 15
## glass fragments given the user-specified arguments
tprob(y, 8:15) 

## produce a summary table for a transfer object
summary(y)

## barplot of transfer probabilities (default)
plot(y, ptype = 0)
plot(y)

## barplot of transfer frequencies
plot(y, ptype = 1)

## histogram
plot(y, ptype = 2)