JMI {JMI}

Jackknife Mutual Information

Description

This function provides method for dependence test. It uses permutation test to determine the rejection region.

Usage

JMI(x, y, BN = 1000)

Arguments

x

n by p sample matrix.

y

n by q sample matrix.

BN

Number of permutations, the default value is 1000.

Value

the output is a list which contains:

• mi: the value of Jackknife Mutual information

• pvalue: the p-value of independence test that based on the permutation of JMI, the value is not provided if BN=0.

References

Zeng, X., Xia, Y., & Tong, H. (2018). Jackknife approach to the estimation of mutual information[J]. Proceedings of the National Academy of Sciences, 201715593.

Examples

 x <- matrix(rnorm(50*3),50,3)
 y <- matrix(rnorm(50*2),50,2)
 #calculate the Jackknife Mutual information between x and y.
 JMI(x,y,0)$mi
 #calculate the p-value of independent test between x and y that based on 500 permutations.
 JMI(x,y,500)$pvalue