gencve (version 0.3)

mae: Mean Absolute Error

Description

This is a widely used criterion since the time of Laplace. Just as least-squares is optimal for mean-square error loss functions, least absolute deviation is optimal for mean absolute error loss functions. See Wikipedia article https://en.wikipedia.org/wiki/Mean_absolute_error.

Usage

mae(yTest, yHat)

Arguments

yTest

test data

yHat

predictions of the test data

Value

mean percentage absolute errors

Details

tba

See Also

gcv, mse, mape, smape

Examples

mape(abs(rnorm(10)), rep(sqrt(2/pi),10))