calc_nmse_from_identity_line: Calculate the normalized mean squared error from the identity...

In richelbilderbeek/gcaer: Interface to GCAE

Calculate the normalized mean squared error from the identity line.

Description

Calculate the normalized mean squared error from the identity line (i.e. the line y = x). If the estimated values are equal to the true values, the MSE is zero.

Usage

calc_nmse_from_identity_line(true_values, estimated_values)

Arguments

true_values	the true values
estimated_values	the estimated values

Value

the mean squared error from the identity line

Author(s)

Richèl J.C. Bilderbeek

See Also

use calc_mse_from_identity_line to calculate the (un-normalized) mean square error from the identity line

Examples

true_values_degrees <- seq(1, 360)
estimated_values_degrees <- true_values_degrees + stats::runif(n = 360)
nmse_degrees <- calc_nmse_from_identity_line(
  true_values = true_values_degrees,
  estimated_values = estimated_values_degrees
)
nmse_degrees

# Convert degrees to radians
true_values_radians <- 2.0 * pi * true_values_degrees / 360
estimated_values_radians <- 2.0 * pi * estimated_values_degrees / 360
nmse_radians <- calc_nmse_from_identity_line(
  true_values = true_values_radians,
  estimated_values = estimated_values_radians
)
nmse_radians

# mse_degrees and mse_radians are the same, due to their normalized scales

richelbilderbeek/gcaer documentation built on March 25, 2024, 3:08 p.m.