assocSignTest {BiCausality}

indpFunc function

Description

This function provides association signs (positive/negative association) inference between i and j. If there is a positive association, it implies i and j trend to have a similar value. For a negative association, however, i and j trend to have an opposite value.

Usage

assocSignTest(mat, i, j, z = c(), alpha = 0.05, IndpThs = 0.05, nboot = 100)

Arguments

mat

is a matrix n by d where n is a number of transactions or samples and d is a number of dimensions.

i

is an ith dimension in mat.

j

is an jth dimension in mat.

z

is a conditioning d-dimensional vector on mat. Given k non-negative-bit positions of z, all k bit positions of samples in the subset of mat must have similar values with these bits.

alpha

is a significance threshold for hypothesis tests (Mann Whitney) that deploys for testing degrees of dependency, association direction, and causal direction. The default is 0.5.

IndpThs

is a threshold for the degree of dependency. In the independence test, to claim that any variables are dependent, the dependency degree must greater than this value significantly. The default is 0.05.

nboot

is a number of bootstrap replicates for bootstrapping deployed to infer confidence intervals and distributions for hypothesis tests. The default is 100.

Value

This function returns results of inference of association signs (positive/negative association) between i and j.

bmean

A mean of sign dependency degrees between variables i and j.

confInv

An alpha*100th percentile confidence interval of sign dependency degrees between variables i and j.

testRes

A Mann-Whitney hypothesis test result for an independence test between variables i and j. The null hypothesis is that the distributions of dependency degrees of i,j differ by a location shift of IndpThs and the alternative is that distributions of dependency degrees of i,j is shifted greater than IndpThs.

Examples

assocSignTest(mat=mat,i=1,j=2)