RMSE: The Root Mean-Square Error(RMSE)

In jjt7549/intervalreg: Regression for Interval-valued Data

Description

The lower bound root mean-square error and the upper bound root mean-square error proposed by Lima Neto and de Carvalho(2008) measure the differences between the predicted values and the observed values.

Usage

RMSE(model)

Arguments

model

imcmuni or imcmtn object, etc..

References

Lima Neto, E.A. and De Carvalho, F.A.T(2010), Constrained linear regression models for symbolic interval-valued variables Computational Statistics and Data Analysis, 54, 333-347

Examples

set.seed(2017)
x1_L = rnorm(30, 3, 0.01) - rnorm(30, 0, 0.01)
x1_U = rnorm(30, 3, 0.01) + rnorm(30, 3, 0.01)
x2_L = runif(30, 1.5, 3) - runif(30, 0, 1)
x2_U = runif(30, 1.5, 3) + runif(30, 1, 2)
y_L = x1_L + x2_L
y_U = x1_U + x2_U
temp <- as.data.frame(cbind(y_L, y_U, x1_L, x1_U, x2_L, x2_U))
m1 <- imcmtn(cbind(y_L, y_U) ~ x1_L + x1_U + x2_L + x2_U, data = temp, b = 100)
RMSE(m1)

jjt7549/intervalreg documentation built on May 19, 2019, 11:40 a.m.