ordPredArgmax: Argmax Mapping from an Estimated Probability Distribution (EPD) to a Predicted Class

Description

Deterministically maps each row of an estimated probability distribution (EPD) matrix to a single predicted class by taking the index of the maximum probability. Rows are normalized to sum to one (within tolerance). Ties can be broken by first, last, or at random among maximizers.

Usage

ordPredArgmax(P, tie_break = c("first", "random", "last"), tol = 1e-12)

Value

An integer vector of length \(n\) with the predicted class indices in \(\{1,\ldots,k\}\) for each row of P.

Arguments

P: A numeric matrix of size \(n \times k\), where each row contains the estimated probabilities \(\hat\pi_{ij}\) for subject \(i\) and classes \(j = 1,\ldots,k\). Values must be nonnegative; rows are normalized to sum to one if needed.
tie_break: Character string indicating how to break ties among equal maxima. One of "first" (default), "last", or "random".
tol: Numeric tolerance used for (i) row-sum checks and (ii) equality when identifying ties among maximum probabilities. Defaults to 1e-12.

Details

The function normalizes each row of P to sum to one (within tol). Rows with (near) zero total probability trigger an error.
If multiple classes achieve the same (within tol) maximum probability, the returned class depends on tie_break:

"first" — smallest index among maximizers (default).
"last" — largest index among maximizers.
"random" — one index sampled uniformly from the set of maximizers.

Examples

Run this code

P <- rbind(
  c(0.05, 0.10, 0.25, 0.60),
  c(0.40, 0.40, 0.10, 0.10), # tie between classes 1 and 2
  c(NA,   0.20, 0.80, 0.00)  # NA treated as 0
)