individual.test: Identify nonzero individual-level partial correlations

Description Usage Arguments Value References See Also Examples

View source: R/individual.test.R

Description

Identify nonzero individual-level partial correlations in time series data by controlling the exceedance rate of the false discovery proportion (FDP) at α=0.05, considering time dependence. Input data X contains values of p interested variables over n periods.

Usage

1
individual.test(indEst, alpha = 0.05, c0 = 0.1, MBT = 3000)

Arguments

indEst

An indEst class object.

alpha

significance level, default value is 0.05.

c0

threshold of the exceedance rate of the false discovery proportion (FDP), default value is 0.1. The choice of c0 depends on the empirical problem. A smaller value of c0 will reduce false positives, but it may also cost more false negatives.

MBT

times of multiplier bootstrap, default value is 3000.

Value

A p*p matrix with values 0 or 1. If the j-th row and k-th column of the matrix is 1, then the partial correlation coefficient between the j-th variable and the k-th variable is identified to be nonzero.

References

Qiu Y. and Zhou X. (2021). Inference on multi-level partial correlations based on multi-subject time series data, Journal of the American Statistical Association, 00, 1-15

See Also

population.test for making inferences on one individual in the population.

Examples

1
2
3
4
## Quick example for the individual-level inference
data(indsim)
pc = individual.est(indsim)       # estimating partial correlation coefficients
Res = individual.test(pc)         # conducting hypothesis test

BrainCon documentation built on Sept. 30, 2021, 5:10 p.m.