entropic_bound {gsaot} | R Documentation |
Entropic lower bounds for entropic optimal transport sensitivity indices
Description
Calculate entropic lower bounds for entropic Optimal Transport sensitivity indices
Usage
entropic_bound(
y,
M,
cost = "L2",
discrete_out = FALSE,
solver = "sinkhorn",
solver_optns = NULL,
scaling = TRUE
)
Arguments
y |
An array or a matrix containing the output values. |
M |
A scalar representing the number of partitions for continuous inputs. |
cost |
(default |
discrete_out |
(default |
solver |
Solver for the Optimal Transport problem. Currently supported options are:
|
solver_optns |
(optional) A list containing the options for the Optimal Transport solver. See details for allowed options and default ones. |
scaling |
(default |
Details
The function allows the computation of the entropic lower bounds.
solver
should be either "sinkhorn"
or "sinkhorn_log"
.
Value
A scalar representing the entropic lower bound.
Examples
N <- 1000
mx <- c(1, 1, 1)
Sigmax <- matrix(data = c(1, 0.5, 0.5, 0.5, 1, 0.5, 0.5, 0.5, 1), nrow = 3)
x1 <- rnorm(N)
x2 <- rnorm(N)
x3 <- rnorm(N)
x <- cbind(x1, x2, x3)
x <- mx + x %*% chol(Sigmax)
A <- matrix(data = c(4, -2, 1, 2, 5, -1), nrow = 2, byrow = TRUE)
y <- t(A %*% t(x))
M <- 25
sink_lb <- entropic_bound(y, M)