shape_aic

Computes the multinomial log-likelihood of observed counts against
four theoretical distributions (uniform, triangular, normal-like,
and exponential/Pareto-like) and returns AIC and DeltaAIC values.

Provides tools for visualizing and analyzing the shape of discrete nominal frequency distributions. The package introduces centered frequency plots, in which nominal categories are ordered from the most frequent category at the center toward less frequent categories on both sides, facilitating the detection of distributional patterns such as uniformity, dominance, symmetry, skewness, and long-tail behavior. In addition, the package supports Pareto charts for the study of dominance and cumulative frequency structure in nominal data. The package is designed for exploratory data analysis and statistical teaching, offering visualizations that emphasize distributional form rather than arbitrary category ordering.

Norberto Asensio

nomiShape

Visualization and Analysis of Nominal Variable Distributions

shape_aic function

<dl><dt>df</dt>
<dd>A data.frame or tibble containing the nominal variable.</dd>
<dt>var</dt>
<dd>Character string giving the name of the nominal variable in <code>df</code>.</dd>
<dt>rate_exp</dt>
<dd>Numeric. Default exponential rate. Only used if tail not clearly exponential.</dd>
<dt>eps</dt>
<dd>Small numeric value added to probabilities to avoid log(0). Default is 1e-12.</dd></dl>

Arguments

Fit Nominal Data to Theoretical Shapes Using AIC (Safe Exponential) — shape_aic

<dl>

<dt>df</dt>
<dd>A data.frame or tibble containing the nominal variable.</dd>


<dt>var</dt>
<dd>Character string giving the name of the nominal variable in <code>df</code>.</dd>


<dt>rate_exp</dt>
<dd>Numeric. Default exponential rate. Only used if tail not clearly exponential.</dd>


<dt>eps</dt>
<dd>Small numeric value added to probabilities to avoid log(0). Default is 1e-12.</dd>

</dl>

shape_aic: Fit Nominal Data to Theoretical Shapes Using AIC (Safe Exponential)

Description

Usage

Value

Arguments