optimize_mse: Optimize MSE of LNC Estimator

In rmi: Mutual Information Estimators

Description

Gaussian process (GP) optimization is used to minimize the MSE of the LNC estimator with respect to the non-uniformity threshold parameter alpha. A normal distribution with compound-symmetric covariance is used as a reference distribution to optimize the MSE of LNC with respect to.

Usage

optimize_mse(rho, N, M, d, k, lower = -10, upper = -1e-10,
  num_iter = 10, init_size = 20, cluster = NULL, verbose = TRUE)

Arguments

rho	Reference correlation.
N	Sample size.
M	Number of replications.
d	Dimension.
k	Neighborhood order.
lower	Lower bound for optimization.
upper	Upper bound for optimization.
num_iter	Number of iterations of GP optimization.
init_size	Number of initial evaluation to estimating GP.
cluster	A parallel cluster object.
verbose	If TRUE then print runtime diagnostic output.

Details

The package tgp is used to fit a treed-GP to the MSE estimates of LNC. A treed-GP is used because the MSE of LNC with respect to alpha exhibits clear non-stationarity. A treed-GP is able to identify the function's different correlation lengths which improves optimization.

rmi documentation built on May 2, 2019, 3:27 a.m.