| tst {copent} | R Documentation |
Two-sample test with copula entropy
Description
Estimating the statistic for two-sample test based on copula entropy.
Usage
tst(s0,s1,n=12,k=3,dt=2)
Arguments
s0, s1 |
two samples with each row as a sample of d-dimensional random variables |
n |
repeat time of estimation to reduce estimation bias, default = 12 |
k |
kth nearest neighbour, default = 3 |
dt |
the type of distance between samples, 1 for Eclidean distance; 2 for Maximum distance |
Details
This program involves estimating the statistic for non-parametric multivariate two-sample test based on copula entropy. It was proposed in Ma (2023). The test statistic is defined as the difference between the copula entropies of the null hypothesis and the alternative of two-sample test.
The argument s0,s1 is for the two samples with each row as a sample of d-dimensional random variables. The argument n is the repeat time of estimation for reducing the estimation bias (dafault = 12). The argument k and dt is used in the kNN method for estimating entropy. k is for the kth nearest neighbour (default = 3) and dt is for the type of distance between samples which has currently two value options (1 for Eclidean distance, and 2(default) for Maximum distance).
Value
The function returns the statistic for two-sample test on s0,s1.
References
Ma, Jian. Two-Sample Test with Copula Entropy. arXiv preprint arXiv:2307.07247, 2023.
Examples
library(mnormt)
rho <- 0.5
sigma <- matrix(c(1,rho,rho,1),2,2)
s0 <- rmnorm(400,c(0,0),sigma)
s1 <- rmnorm(500,c(5,5),sigma)
tst(s0,s1)