Analyzes pattern causality matrices to classify the nature of causality between variables. This function provides core functionality for pattern causality analysis and can be used both independently and as part of larger analysis workflows.
Arguments
- PC
Three-dimensional array; pattern causality matrices
- dur
Numeric vector; time points to analyze
- hashedpatterns
Numeric vector; pattern identifiers
- X
Numeric vector; reference for output length
- weighted
Logical; if TRUE, uses weighted causality strength
- verbose
Logical; if TRUE, prints computation details
Value
A pc_nature object containing:
no_causality: Vector of no causality strengths
positive: Vector of positive causality strengths
negative: Vector of negative causality strengths
dark: Vector of dark causality strengths
Details
Nature of Causality Analysis
This function analyzes the structure of pattern causality matrices to determine four types of causality:
No Causality: When no significant relationship is detected
Positive Causality: When patterns show positive influence
Negative Causality: When patterns show negative influence
Dark Causality: When patterns show complex or indirect influence
See also
pcLightweight for basic causality analysis
pcFullDetails for detailed analysis
pcMatrix for causality matrix computation
Examples
# \donttest{
# Generate example data
PC <- array(runif(27), dim = c(3,3,3))
dur <- 1:3
hashedpatterns <- 1:3
X <- rnorm(10)
# Analyze causality nature
result <- natureOfCausality(PC, dur, hashedpatterns, X)
print(result)
#> Pattern Causality Nature Analysis
#> --------------------------------
#> No Causality: 0
#> Positive Causality: 0
#> Negative Causality: 0
#> Dark Causality: 3
# }