A statistical simulation method for causality inference of Boolean variables

Abstract

In present study, a statistical simulation method for causality inference of Boolean variables was proposed. First, I used statistical simulation to generate artificial data of two Boolean variables with known independent and dependent variables. A law was drawn from the simulation analysis of the artificial data. For a set of data of two Boolean variables, a randomization method was proposed and used to test the statistical significance of the Boolean correlation measure (point correlation, quartile correlation, or Jaccard correlation, etc.). The causality inference was then conducted to observed data based on the law. Finally, the statistical simulation was used to determine the statistic significance of the causality. Full Matlab codes were presented also.

Keywords

• causality
• inference
• correlation
• point correlation
• jaccard
• boolean variables
• randomization
• statistical simulation
• machine learning
• artificial intelligence
• independent variable
• causal inference