Skip to contents

Relative Absolute Error (RAE)

Source: R/reg_RAE.R
RAE.Rd

It estimates the RAC for a continuous predicted-observed dataset.

Usage

RAE(data = NULL, obs, pred, tidy = FALSE, na.rm = TRUE)

Arguments

data

(Optional) argument to call an existing data frame containing the data.

obs

Vector with observed values (numeric).

pred

Vector with predicted values (numeric).

tidy

Logical operator (TRUE/FALSE) to decide the type of return. TRUE returns a data.frame, FALSE returns a list; Default : FALSE.

na.rm

Logic argument to remove rows with missing values (NA). Default is na.rm = TRUE.

Value

an object of class numeric within a list (if tidy = FALSE) or within a data frame (if tidy = TRUE).

Details

The RAE normalizes the Mean Absolute Error (MAE) with respect to the total absolute error. It is calculated as the ratio between the sum of absolute residuals (error of predictions with respect to observations) and the sum of absolute errors of observations with respect to its mean. It presents its lower bound at 0 (perfect fit), and has no upper bound. It can be used to compare models using different units. For the formula and more details, see online-documentation

Examples

# \donttest{
set.seed(1)
X <- rnorm(n = 100, mean = 0, sd = 10)
Y <- X + rnorm(n=100, mean = 0, sd = 3)
RAE(obs = X, pred = Y)
#> $RAE
#> [1] 0.3202058
#> 
# }