delta_k(x) = -log|S_k| / 2 - (x - mu_k)' S_k^-1 (x - mu_k) / 2 + log pi_k with
per-class covariances (divisor n_k - 1) (ESL eq 4.12).
Usage
morie_esl_qda(X, y, query = NULL)
Arguments
- X
Training predictors, N by p.
- y
Class labels.
- query
Points to score (default X).
Value
Named list: estimate, prediction, discriminants, classes, priors,
log_dets, n, p, K.
References
Hastie, Tibshirani & Friedman (2009), sec. 4.3.
Examples
i <- 1:30
X <- cbind(sin(i), cos(3 * i) + (i %% 3))
morie_esl_qda(X, i %% 3, rbind(c(0, 1)))$prediction
#> [1] 1
