Exploratory.path.analysis: Temporary title

In BillShipley/CauseAndCorrelation: Functions for Path Analysis and SEM

Temporary title

Description

This function impliments the CI (Causal Inference) Algorithm of Pearl (2009). It takes multivariate observations (3 or more variables) as input, determines the partly directed dependency graph that corresponds to the patterns of correlation and partial correlation in the input data, and then attempts to completely orient it to produce a d-separation equivalent DAG. It then tests this DAG against the data using the chosen significance level. The algorithm begins with a severe requirement for d-separation (alpha=0.01) and increases this incrimentally, up to the value specified in "upper.bound" and stops when it finds a DAG that does not contradict the data.

Usage

Exploratory.path.analysis(dat, upper.bound = 0.5, significance.level = 0.05) 

Arguments

dat	a matrix or data frame containing only numeric variables
upper.bound	the highest alpha significance level used to determin if a d-seperation claim is true or not
significance.level	the significance level used to test the SAG

Value

A partially oriented graph and an equivalent DAG, along with the null probability

Author(s)

Bill Shipley

References

Pearl, J. 2009. Causality. Models, Reasoning , and Inference (2nd edition). Cambridge University Press. Shipley, B. 2016. Cause and Correlation in Biology: A user's guide to path analysis, structural equations and causal inference in R. Cambridge University Press.

See Also

Causal.Inference

Examples

set.seed(123)
dat <- gen.data()
Exploratory.path.analysis(dat)

BillShipley/CauseAndCorrelation documentation built on Jan. 31, 2023, 4:20 a.m.