BLB.multi: BLB for multivariate data

In EllaKaye/Module1:

Description

BLB function that computes the (1-alpha)% CI and the standard errors of the parameter estimate components

Usage

BLB.multi(data, gamma = 0.7, s = 20, r = 100, lambda = 10^(-5),
  alpha = 0.05)

Arguments

data	a matrix or dataframe
gamma	specifies the size of the subsample by b=n^gamma
s	number of subsamples
r	number of resamples per subsample
lambda	is the penalty in L2 norm for the ridge regression
alpha	the level of the confidence for CI, respecified to 0.05

Value

a list consisting of

output_CI	average confidence intervals for the parameter estimate components
output_se	average standard errors for the parameter estimates components

Examples

n <- 100
d <- 5
X <- mvrnorm(n, mu = rep(0, d) ,Sigma <- diag(d))
epsilon <- rnorm(n, mean = 0, sd = sqrt(10))
t_theta <- as.matrix(rep(1, d))
Y <- X %*% t_theta + epsilon
data <- cbind(X, Y)
my_result <- BLB.multi(data, 0.7, 15, 100, 10^(-5))

EllaKaye/Module1 documentation built on May 6, 2019, 3:25 p.m.