mr2: Multinomial R-square

In VMTdeJong/mPerformance: Performance measures for multinomial outcome data

Description

Compute various R-square measures: Cox, Nagelkerke and McFadden. Requires the predicted (or fitted) probability matrix p, and one of the following: labels, indices or indicator.matrix. Preferably one of the two former.

Usage

mr2(p, labels, indices, indicator.matrix, names = colnames(p), na.rm = T)

Arguments

p	An n x K matrix of probabilities, where n is the number of observations, and K the number of mutually exclusive outcome categories.
labels	Vector of length n, containing the labels (character or factor) of the observed outcome categories. If specified, must correspond with the column names of p or with names.
indices	Optional. A vector of length n, containing the indices k, k = 1,...,K, of the observed outcome categories. If specified, these indices must corresond with their respective indices in p.
indicator.matrix	Optional. An n x K matrix indicating the outcome category of each observation, where n is the number of observations, and K the number of mutually exclusive outcome categories. If specified, the order of the columns should correspond with the order of the columns of p.
names	Optional. What are the labels to which the columns of p should be matched? By default, the colnames of the outcome matrix p.
na.rm	logical. Should missing values (including NaN) be removed?

Value

mr2 provides a data.frame of R-square values by the methods of Cox, Nagelkerke and Mcfadden.

References

Nagelkerke NJ. A note on a general definition of the coefficient of determination. Biometrika. 1991 Sep 1;78(3):691-2. McFadden D. Conditional logit analysis of qualitative choice behavior.

Examples

# If we observe outcomes A, B and C:
labels <- c("A", "B", "c")
# The fitted probabilities of an intercept only model are given by 1/3:
probabilities <- matrix(1/3, nrow = 3, ncol = 3)
colnames(probabilities) <- labels
# Then the multinomial R-squares can be obtained with:
mr2(probabilities, labels)
# Or:
mr2(probabilities, as.factor(labels))
# Similary, we can use the indices of the observed outcome
# categories:
mr2(probabilities, indices = c(1,2,3))

VMTdeJong/mPerformance documentation built on May 14, 2019, 7:42 a.m.