bnlearn (version 3.4)

ci.test: Independence and Conditional Independence Tests

Description

Perform either an independence test or a conditional independence test.

Usage

## S3 method for class 'character':
ci.test(x, y = NULL, z = NULL, data, test = NULL, B = NULL, debug = FALSE, ...)
## S3 method for class 'data.frame':
ci.test(x, test = NULL, B = NULL, debug = FALSE, ...)
## S3 method for class 'numeric':
ci.test(x, y = NULL, z = NULL, test = NULL, B = NULL, debug = FALSE, ...)
## S3 method for class 'factor':
ci.test(x, y = NULL, z = NULL, test = NULL, B = NULL, debug = FALSE, ...)
## S3 method for class 'default':
ci.test(x, ...)

Arguments

x

a character string (the name of a variable), a data frame, a numeric vector or a factor object.

y

a character string (the name of another variable), a numeric vector or a factor object.

z

a vector of character strings (the names of the conditioning variables), a numeric vector, a factor object or a data frame. If NULL an independence test will be executed.

data

a data frame containing the variables to be tested.

test

a character string, the label of the conditional independence test to be used in the algorithm. If none is specified, the default test statistic is the mutual information for categorical variables, the Jonckheere-Terpstra test f

B

a positive integer, the number of permutations considered for each permutation test. It will be ignored with a warning if the conditional independence test specified by the test argument is not a permutation test.

debug

a boolean value. If TRUE a lot of debugging output is printed; otherwise the function is completely silent.

...

extra arguments from the generic method (currently ignored).

Value

• An object of class htest containing the following components:
• statistic
  the value the test statistic.
• parameter
  the degrees of freedom of the approximate chi-squared or t distribution of the test statistic; the number of permutationscomputed by Monte Carlo tests. Semiparametric tests have both.
• p.value
  the p-value for the test.
• method
  a character string indicating the type of test performed, and whether Monte Carlo simulation or continuity correction was used.
• data.name
  a character string giving the name(s) of the data.
• null.value
  the value of the test statistic under the null hypothesis, always 0.
• alternative
  a character string describing the alternative hypothesis

References

for parametric and discrete permutation tests:

Edwards DI (2000). Introduction to Graphical Modelling. Springer, 2nd edition.

for shrinkage tests:

Hausser J, Strimmer K (2009). "Entropy inference and the James-Stein estimator, with application to nonlinear gene association networks". Statistical Applications in Genetics and Molecular Biology, 10, 1469-1484.

Ledoit O, Wolf M (2003). "Improved Estimation of the Covariance Matrix of Stock Returns with an Application to Portfolio Selection". Journal of Empirical Finance, 10, 603-621.

for continuous permutation tests:

Legendre P (2000). "Comparison of Permutation Methods for the Partial Correlation and Partial Mantel Tests". Journal of Statistical Computation and Simulation, 67, 37-73.

for semiparametric discrete tests:

Tsamardinos I, Borboudakis G (2010). "Permutation Testing Improves Bayesian Network Learning". In "Machine Learning and Knowledge Discovery in Databases", pp. 322-337. Springer.

See Also

choose.direction, arc.strength.

Examples

data(gaussian.test)
data(learning.test)

# using a data frame and column labels.
ci.test(x = "F" , y = "B", z = c("C", "D"), data = gaussian.test)
# using a data frame.
ci.test(gaussian.test)
# using factor objects.
attach(learning.test)
ci.test(x = F , y = B, z = data.frame(C, D))