witness.optim: General purpose fitting for WITNESS

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

Description

optim.witness uses a steepest-descent algorithm to fit empirical data using the WITNESS model

Usage

  witness.optim(param.form, data.set, generate.new = TRUE,
    ...)

Arguments

param.form	A parameters matrix. See details.
data.set	The dataset to be fit. Columns correspond to Target Chosen, Foil Chosen, and Lineup Rejected. Rows are for each experiment.
generate.new	If starting values are not provided, making this TRUE will generate new starting values.
...	Other parameters passed to the WITNESS model

Details

The user must specify the parameters of the model in matrix form. The columns of the matrix are as follows: e (encoding parameter), ssp (similarity of suspect to perp), sfs (similarity of foil to suspect), c (decision criteria), and wa (relative vs. absolute criteria). The rows correspond to each data row to be fit. See examples.

Value

This function returns two objects:

fitted	A matrix in the same form as param.form that contains the fitted values.
fit	The rmse of the fitted object.

See Also

witness.genetic, witness.est

Examples

# create an matrix with data to be fit
dataMatrix = matrix(c(.471, .230, .350,
							.208, .137, .513,
							.396, .431, .242,
							.166, .081, .669), nrow=4, byrow=TRUE)
# specify 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)
# find optimal parameters (currently commented to save time)
# fit = witness.optim(params, data.set=dataMatrix, sample.size=100, meth="WITC")
# fit

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