fftrees_threshold_numeric_grid

Perform a grid search over thresholds and return accuracy statistics for a given numeric cue

Create, visualize, and test fast-and-frugal decision trees (FFTs) using the algorithms and methods described in Phillips, Neth, Woike & Gaissmaier (2017).
FFTs are simple and transparent decision trees for solving binary classification problems.
FFTs can be preferable to more complex algorithms because they require very little information, are easy to communicate, and are robust against overfitting.

Hansjoerg Neth

FFTrees

Generate, Visualise, and Evaluate Fast-and-Frugal Decision Trees

Nathaniel Phillips

Jan Woike

Wolfgang Gaissmaier

fftrees_threshold_numeric_grid function

<dl><dt>thresholds</dt>
<dd>numeric. A vector of thresholds to consider.</dd>
<dt>cue_v</dt>
<dd>numeric. Feature values.</dd>
<dt>criterion_v</dt>
<dd>logical. A logical vector of (TRUE) criterion values.</dd>
<dt>directions</dt>
<dd>character. Possible directions to consider.</dd>
<dt>sens.w</dt>
<dd>numeric. Sensitivity weight parameter (from 0 to 1, for computing <code>wacc</code>).
Default: <code>sens.w = .50</code>.</dd>
<dt>cost.each</dt>
<dd>numeric. Cost to add to each value (e.g.; cost of the cue).</dd>
<dt>cost.outcomes</dt>
<dd>list. A list of length 4 with names 'hi', 'fa', 'mi', and 'cr' specifying
the costs of a hit, false alarm, miss, and correct rejection, respectively.
For instance, <code>cost.outcomes = listc("hi" = 0, "fa" = 10, "mi" = 20, "cr" = 0)</code> means that
a false alarm and miss cost 10 and 20, respectively, while correct decisions have no cost.</dd>
<dt>goal.threshold</dt>
<dd>character. A string indicating the statistic to maximize when calculting cue thresholds: "acc" = overall accuracy, "wacc" = weighted accuracy, "bacc" = balanced accuracy.</dd></dl>

Arguments

Perform a grid search over thresholds and return accuracy statistics for a given numeric cue — fftrees_threshold_numeric_grid

<dl>

<dt>thresholds</dt>
<dd>numeric. A vector of thresholds to consider.</dd>


<dt>cue_v</dt>
<dd>numeric. Feature values.</dd>


<dt>criterion_v</dt>
<dd>logical. A logical vector of (TRUE) criterion values.</dd>


<dt>directions</dt>
<dd>character. Possible directions to consider.</dd>


<dt>sens.w</dt>
<dd>numeric. Sensitivity weight parameter (from 0 to 1, for computing <code>wacc</code>).
Default: <code>sens.w = .50</code>.</dd>


<dt>cost.each</dt>
<dd>numeric. Cost to add to each value (e.g.; cost of the cue).</dd>


<dt>cost.outcomes</dt>
<dd>list. A list of length 4 with names 'hi', 'fa', 'mi', and 'cr' specifying
the costs of a hit, false alarm, miss, and correct rejection, respectively.
For instance, <code>cost.outcomes = listc("hi" = 0, "fa" = 10, "mi" = 20, "cr" = 0)</code> means that
a false alarm and miss cost 10 and 20, respectively, while correct decisions have no cost.</dd>


<dt>goal.threshold</dt>
<dd>character. A string indicating the statistic to maximize when calculting cue thresholds: "acc" = overall accuracy, "wacc" = weighted accuracy, "bacc" = balanced accuracy.</dd>

</dl>

fftrees_threshold_numeric_grid: Perform a grid search over thresholds and return accuracy statistics for a given numeric cue

Description

Usage

Value

Arguments

See Also