plotCI: Plot hyperrectangular and ellipsoidal confidence intervals

In cmottet/GLP: GLP

Description

Plot hyperrectangular and ellipsoidal confidence intervals

Usage

plotCI(sample, CI)

Arguments

sample	Vector of sample points
CI	List containing either hyperrectangle, data.frame containing the lower bounds and upper bounds of the derivatives and truncated moments., or \itemellipsoid, a list containing the vector of means, the covariance matrix, and the radius of the ellipsoid describing the confidence interval

Value

an object of class ggplot2

Examples

set.seed(100)
n <- 300 ;  mu <- c(1,2,3) ; Sigma <- matrix(c(6,0,0,0,1,1,0,1,1),ncol =3) ; alpha = 0.05
sample <- MASS::mvrnorm(n,mu,Sigma) %>% as.data.frame
CIellipsoid <- list(mu = colMeans(sample, na.rm = TRUE), Sigma = cov(sample, use = "complete.ob"), radius = sqrt(qchisq(1-alpha/2, df = 2)))
CIrectangular <- apply(sample, 2, quantile, probs = c(alpha/2,1-alpha/2))
CI <- list(ellipsoid = CIellipsoid, hyperrectangle = CIrectangular)
plotCI(sample,CI)

n <- 300 ;  mu <- c(1,2) ; Sigma <- matrix(c(6,1,1,1),ncol =2) ; alpha = 0.05
sample <- MASS::mvrnorm(n,mu,Sigma) %>% as.data.frame
CIellipsoid <- list(mu = colMeans(sample, na.rm = TRUE), Sigma = cov(sample, use = "complete.ob"), radius = sqrt(qchisq(1-alpha/2, df = 2)))
CIrectangular <- apply(sample, 2, quantile, probs = c(alpha/2,1-alpha/2))
CI <- list(ellipsoid = CIellipsoid, hyperrectangle = CIrectangular)
plotCI(sample,CI)

cmottet/GLP documentation built on May 6, 2019, 12:05 a.m.