loglikcub00

Vector of the absolute frequency distribution

freq

Logit transform of uncertainty parameter

beta0

Logit transform of fFeeling parameter

gama0

Compute the log-likelihood function of a CUB model without covariates for a given 
absolute frequency distribution.

internal

For ordinal rating data, consider the accelerated Expectation-Maximization algorithm to estimate and test models within the family of CUB models (where CUB stands for Combination of a discrete Uniform and a shifted Binomial distributions). The procedure is built upon Louis' identity for the observed information matrix. Best-subset variable selection for CUB regression models is then implemented on such basis. The methods here implemented are illustrated and discussed in the preprint available from Researchgate by Simone R. (2020) <https://tinyurl.com/vvk563e>.

Rosaria Simone

FastCUB

Fast EM and Best-Subset Selection for CUB Models for Rating Data

loglikcub00 function

<dl><dt>m</dt>
<dd>Number of ordinal categories</dd>
<dt>freq</dt>
<dd>Vector of the absolute frequency distribution</dd>
<dt>beta0</dt>
<dd>Logit transform of uncertainty parameter</dd>
<dt>gama0</dt>
<dd>Logit transform of fFeeling parameter</dd></dl>

Arguments

Log-likelihood function of a CUB model without covariates — loglikcub00

<dl>

<dt>m</dt>
<dd>Number of ordinal categories</dd>


<dt>freq</dt>
<dd>Vector of the absolute frequency distribution</dd>


<dt>beta0</dt>
<dd>Logit transform of uncertainty parameter</dd>


<dt>gama0</dt>
<dd>Logit transform of fFeeling parameter</dd>

</dl>

loglikcub00: Log-likelihood function of a CUB model without covariates

Description

Usage

Arguments