witness-package: This package fits eyewitness data using Clark's (2003)...

In dustinfife/witness: This package fits eyewitness data using Clark's (2003) WITNESS model

Description

WITNESS is a computational memory model for eyewitness identification. This package is able to estimate probabilities based on user-inputted parameters as well as find optimal parameters using a steepest-descent and/or a genetic algorithm.

Details

Package:	witness
Type:	Package
Version:	1.0
Date:	2012-10-25
License:	GNU 2

Author(s)

Maintainer: Dustin Fife <[e-mail]>

References

Clark, S. E. (2003). A memory and decision model for eyewitness identification. Applied Cognitive Psychology, 17, 629-654.

See Also

NONE

Examples

# create data matrix
dataMatrix = matrix(c(.471, .230, .350,
							.208, .137, .513,
							.396, .431, .242,
							.166, .081, .669), nrow=4, byrow=TRUE)
# create an parameter.form
params = matrix(c(rep("e", times=4),
		0, 0, 0, 0,
		"sfs1", "sfs1", "sfs2", "sfs2",
		"cr1", "cr2", "cr1", "cr2",
		1, 1, 1, 1), nrow=4)


est = witness.est(parameters=params, data=dataMatrix, meth="WITC", sample.size=100)
est

dustinfife/witness documentation built on May 15, 2019, 6:05 p.m.