predict.serp

This function takes a fitted <code>serp</code> object produced by serp() and
produces predicted values. Type of predictions returned include response,
link and class. Prediction is also possible with new set of values having
the same column names as in the original values used for the model fit.

Implements a regularization method for cumulative link models using the
Smooth-Effect-on-Response Penalty (SERP). This method allows flexible
modeling of ordinal data by enabling a smooth transition from a general
cumulative link model to a simplified version of the same model. As the
tuning parameter increases from zero to infinity, the subject-specific
effects for each variable converge to a single global effect.
The approach addresses common issues in cumulative link models, such as
parameter unidentifiability and numerical instability, by maximizing a
penalized log-likelihood instead of the standard non-penalized version.
Fitting is performed using a modified Newton's method. Additionally, the
package includes various model performance metrics and descriptive tools.
For details on the implemented penalty method, see
Ugba (2021) <doi:10.21105/joss.03705> and
Ugba et al. (2021) <doi:10.3390/stats4030037>.

Ejike Ugba

serp

Smooth Effects on Response Penalty for CLM

Ejike R. Ugba

predict.serp function

<dl><dt>object</dt>
<dd>An object of class <code>serp</code>.</dd>
<dt>type</dt>
<dd>could be any of these: response, link or terms.</dd>
<dt>newdata</dt>
<dd>fresh dataset with all relevant variables.</dd>
<dt>...</dt>
<dd>additional arguments.</dd></dl>

Arguments

Prediction from fitted serp model — predict.serp

<dl>

<dt>object</dt>
<dd>An object of class <code>serp</code>.</dd>


<dt>type</dt>
<dd>could be any of these: response, link or terms.</dd>


<dt>newdata</dt>
<dd>fresh dataset with all relevant variables.</dd>


<dt>...</dt>
<dd>additional arguments.</dd>

</dl>

predict.serp: Prediction from fitted serp model

Description

Usage

Value

Arguments

See Also

Examples