predict_qda function

Quadratic discrimination prediction

The function uses the output from the function qda

(Section A.3.2) and a $P$-vector $X$, and calculates the predicted group for this $X$.

predict_qda(qd, newx)

Arguments

• qd: The output from qda.
• newx: A $P$-vector $X$ whose components match the variables used in the qda function.

Returns

A $K$-vector of the discriminant values $d_k^Q(X)$ in (11.48) for the given $X$.

Examples

# Load Iris Data
data(iris)

# Build data
x.iris <- as.matrix(iris[, 1:4])
n <- nrow(x.iris)
# Gets group vector (1, ... , 1, 2, ..., 2, 3, ... , 3)
y.iris <- rep(1:3, c(50, 50, 50))

# Perform QDA
qd.iris <- qda(x.iris, y.iris)
yhat.qd <- NULL
for (i in seq_len(n)) {
  yhat.qd <- c(yhat.qd, imax(predict_qda(qd.iris, x.iris[i, ])))
}
table(yhat.qd, y.iris)

See Also

qda, imax