LL_baseline: Baseline category logistic regression log-likelihood

In gspcr: Generalized Supervised Principal Component Regression

Baseline category logistic regression log-likelihood

Description

Computes the baseline category logistic regression log-likelihood given a nominal categorical variable and the corresponding GLM linear predictor values.

Usage

LL_baseline(y, x, mod)

Arguments

y	factor or disjunctive table representation recording a nominal variable with 3 or more categories.
x	data.frame (or matrix) containing predictor values.
mod	multinom object containing the estimated baseline-category logit model.

Details

If x and y are equal to the data on which mod has been trained, this function returns the same result as the default logLink function. If x and y are new, the function returns the log-likelihood of the new data under the trained model.

A disjunctive table is a matrix representation of a multi-categorical variable. The dimensionality of the matrix is i times j, with i = number of observations, and j = number of categories. y_{ij} is equal to 1 if observation i responded with category j, and it is equal to 0 otherwise. The log-likelihood equation is based on Agresti (2002, p. 192).

Value

A list containing:

• ll atomic vector of length 1 containing the log-likelihood value.

• sc numeric matrix containing the systematic component for the input x and mod.

Author(s)

Edoardo Costantini, 2023

References

Agresti, A. (2012). Categorical data analysis (Vol. 792). John Wiley & Sons.

gspcr documentation built on May 29, 2024, 2:44 a.m.