ggm (version 2.2)

pcor.test: Test for zero partial association

Description

Test for conditional independence between two variables, given the other ones, assuming a multivariate normal distribution.

Usage

pcor.test(r, q, n)

Arguments

r

a partial correlation coefficient, computed by pcor.

q

the number of variables in the conditioning set.

n

integer > 0, the sample size.

Value

• tval
  The Student's t-test statistic.
• df
  The degrees of freedom
• pvalue
  The P-value, assuming a two-sided alternative.

See Also

pcor, shipley.test

Examples

## Are 2,3 independent given 1?
data(marks)
pcor.test(pcor(c(2,3,1), var(marks)), 1, n=88)