AICc

<p>The Akaike information criterion corrected for small sample size is a measure of the relative quality of a model.  The AICc is calculated from a 'dark' object. </p>

models

regression

utilities

math

The recovery of visual sensitivity in a dark environment is known
as dark adaptation. In a clinical or research setting the recovery is typically
measured after a dazzling flash of light and can be described by the Mahroo,
Lamb and Pugh (MLP) model of dark adaptation. The functions in this package take
dark adaptation data and use nonlinear regression to find the parameters of the
model that 'best' describe the data. They do this by firstly, generating rapid
initial objective estimates of data adaptation parameters, then a multi-start
algorithm is used to reduce the possibility of a local minimum. There is also a
bootstrap method to calculate parameter confidence intervals. The functions rely
upon a 'dark' list or object. This object is created as the first step in the
workflow and parts of the object are updated as it is processed.

Dr MF Kelly

Dark

The Analysis of Dark Adaptation Data

AICc function

<dl>
 <dt>obj</dt>
<dd><p>A dark object

This object must have at least the following elements:</p><table class="table">
<tr><td>obj$time</td><td>to calculate the number of observations</td></tr>
<tr><td>obj$Pn</td><td>the number of parameters in the model</td></tr>
<tr><td>obj$val</td><td>the sum of squared residual error</td></tr>
</table><p></p><p></p></dd><p></p><p></p></dl>

Arguments

<p>
Jeremiah MF Kelly</p>
<p>Mumac Ltd, SK7 6NR, GB</p>

Author

Akaike information criterion 
 — AICc

<dl>
 <dt>obj</dt>
<dd><p>A dark object

This object must have at least the following elements:</p><table class='table'>
<tr><td>obj$time</td><td>to calculate the number of observations</td></tr>
<tr><td>obj$Pn</td><td>the number of parameters in the model</td></tr>
<tr><td>obj$val</td><td>the sum of squared residual error</td></tr>
</table>

</dd>

</dl>

AICc: Akaike information criterion

Description

Usage

Value

Arguments

Author

References

See Also

Examples

obj$time	to calculate the number of observations
obj$Pn	the number of parameters in the model
obj$val	the sum of squared residual error