Type:	Package
Title:	Pattern Causality Algorithm
Version:	0.2.1
Maintainer:	Hui Wang <huiw1128@gmail.com>
Description:	A comprehensive package for detecting and analyzing causal relationships in complex systems using pattern-based approaches. Key features include state space reconstruction, pattern identification, and causality strength evaluation.
License:	GPL-3 | file LICENSE
Depends:	R (≥ 4.1.0)
Imports:	stats, plot3D, ggplot2, reshape2, grid, ggthemes, tidyr, statebins, ggrepel, RColorBrewer, scales, gridExtra, parallel
Suggests:	knitr, rmarkdown, stringr, testthat (≥ 3.0.0), lintr, zoo, graphics,
Encoding:	UTF-8
LazyData:	true
Config/Needs/website:	tidyverse/tidytemplate
URL:	https://github.com/skstavroglou/pattern_causality/
BugReports:	https://github.com/skstavroglou/pattern_causality/issues
Config/testthat/edition:	3
RoxygenNote:	7.3.2
VignetteBuilder:	knitr
NeedsCompilation:	no
Packaged:	2025-02-15 12:31:57 UTC; huiwang
Author:	Stavros Stavroglou [aut], Athanasios Pantelous [aut], Hui Wang [aut, cre]
Repository:	CRAN
Date/Publication:	2025-02-15 13:40:09 UTC

Illapel Ecological Dataset

Description

Raw rodent and rainfall data collected from the Las Chinchillas National Reserve near Illapel, Coquimbo Region of Chile. This dataset provides ecological time series for studying species interactions and environmental effects.

Usage

AUCO

Format

A data frame with rodent and rainfall data.

Details

Illapel Ecological Dataset

Source

Las Chinchillas National Reserve Research Station

Examples

data(AUCO)
head(AUCO)
summary(AUCO)

Dow Jones Stock Price Dataset

Description

A comprehensive dataset containing daily stock prices for 29 companies listed in the Dow Jones Industrial Average (DJIA). The dataset includes opening, closing, high, and low prices for each stock.

Usage

DJS

Format

A data frame with daily stock prices for 29 companies.

Details

Dow Jones Stock Price Dataset

Source

Yahoo Finance

Examples

data(DJS)
head(DJS)
summary(DJS)

Climate Indices Dataset

Description

A comprehensive time series dataset containing various climate indices used for pattern causality analysis. This dataset includes multiple climate indicators measured over time.

Usage

climate_indices

Format

A data frame with 100 rows and 5 columns:

Date

Date; Date of the measurement

AO

Numeric; Arctic Oscillation index

AAO

Numeric; Antarctic Oscillation index

NAO

Numeric; North Atlantic Oscillation index

PNA

Numeric; Pacific/North American index

Details

Climate Indices Dataset

Source

https://www.cpc.ncep.noaa.gov/

Examples

data(climate_indices)
head(climate_indices)
summary(climate_indices)

Distance Metric Interface

Description

A generic interface for computing distances between observations using either built-in or custom distance metrics.

Usage

distanceMetric(x, method = "euclidean", ...)

## Default S3 method:
distanceMetric(x, method = "euclidean", ...)

## S3 method for class 'custom'
distanceMetric(x, method, ...)

Arguments

Details

Generic Interface for Distance Metrics

Value

A distance object or matrix containing pairwise distances

Methods (by class)

• distanceMetric(default): Default method using stats::dist

• distanceMetric(custom): Custom distance metric implementation

Examples

## Not run: 
# Using default method
x <- matrix(rnorm(100), ncol=2)
d1 <- distanceMetric(x, "euclidean")

# Using custom method
custom_dist <- function(x) as.dist(crossprod(x))
d2 <- distanceMetric(x, method=custom_dist)

## End(Not run)

Search for Optimal Parameters in Pattern Causality Analysis

Description

Searches for the optimal embedding dimension (E) and time delay (tau) to maximize the accuracy of causality predictions in a dataset. This function implements a grid search approach to evaluate different parameter combinations.

Usage

optimalParametersSearch(
  Emax,
  tauMax,
  metric = "euclidean",
  distance_fn = NULL,
  state_space_fn = NULL,
  dataset,
  h = 0,
  weighted = FALSE,
  relative = TRUE,
  verbose = FALSE
)

Arguments

Details

Search for Optimal Parameters in Pattern Causality Analysis

This function evaluates each combination of embedding dimension and time delay for their effectiveness in detecting different types of causality:

• Total causality: Overall causal relationship strength

• Positive causality: Direct positive influences

• Negative causality: Direct negative influences

• Dark causality: Complex or indirect causal relationships

Value

A pc_params object containing:

• accuracy_summary: A data frame summarizing the accuracy for each parameter combination.

• computation_time: The time taken for the analysis.

• parameters: A list of the input parameters used.

Examples

data(climate_indices)
dataset <- climate_indices[, -1]
optimalParams <- optimalParametersSearch(
  Emax = 3, 
  tauMax = 3, 
  metric = "euclidean",
  dataset = dataset,
  h = 1,
  weighted = FALSE
)
print(optimalParams)

Calculate Pattern Causality Accuracy

Description

Evaluates the causality prediction accuracy across multiple time series within a dataset using the PC Mk. II Light method. This function analyzes pairwise causality relationships and computes different types of causality measures.

Usage

pcAccuracy(
  dataset,
  E,
  tau,
  metric = "euclidean",
  h,
  weighted,
  distance_fn = NULL,
  state_space_fn = NULL,
  relative = TRUE,
  verbose = FALSE
)

Arguments

Details

Calculate Pattern Causality Accuracy

Value

An object of class "pc_accuracy" containing:

• parameters: List of input parameters (E, tau, metric, h, weighted)

• total: Mean total causality across all pairs

• positive: Mean positive causality across all pairs

• negative: Mean negative causality across all pairs

• dark: Mean dark causality across all pairs

• matrices: Raw causality matrices for each type

See Also

pcMatrix for analyzing individual causality matrices pcLightweight for pairwise causality analysis

Examples

data(climate_indices)
data <- climate_indices[, -1]
results <- pcAccuracy(dataset = data, E = 3, tau = 1, 
                     metric = "euclidean", h = 1, 
                     weighted = TRUE, verbose = TRUE)
print(results)

Cross Pattern Causality Matrix Analysis

Description

Analyzes pattern causality relationships between multiple time series in X and multiple time series in Y by computing pairwise causality measures and organizing them into a matrix.

Usage

pcCrossMatrix(
  X,
  Y,
  E,
  tau,
  metric = "euclidean",
  h,
  weighted = TRUE,
  distance_fn = NULL,
  state_space_fn = NULL,
  relative = TRUE,
  verbose = FALSE,
  n_cores = 1
)

Arguments

Details

Compute Cross Pattern Causality Matrix Analysis

The function performs these key steps:

• Validates input data and parameters

• Computes pairwise causality measures between X and Y

• Organizes results into a causality matrix

• Provides summary statistics for each causality type

Value

A pc_matrix object containing causality matrices

Related Packages

• vars: Vector autoregression analysis

• tseries: Time series analysis tools

• forecast: Time series forecasting methods

Pattern Causality Cross-Validation Analysis

Description

Evaluates the robustness of pattern causality measures through repeated sampling analysis. This function performs cross-validation by analyzing multiple subsets of the data to assess the stability of causality relationships.

Usage

pcCrossValidation(
  X,
  Y,
  E,
  tau,
  metric = "euclidean",
  h,
  weighted,
  distance_fn = NULL,
  state_space_fn = NULL,
  numberset,
  random = TRUE,
  bootstrap = 1,
  verbose = FALSE,
  n_cores = 1,
  relative = TRUE
)

Arguments

Details

Perform Pattern Causality Cross-Validation Analysis

The function implements these key steps:

• Validates input parameters and data

• Performs stratified sampling of time series data

• When random=TRUE and bootstrap>1, performs bootstrap sampling

• Computes pattern causality measures for each sample

• Aggregates results across all samples

When bootstrap sampling is enabled (random=TRUE and bootstrap>1), the function returns statistics including mean, 5% quantile, 95% quantile, and median for each sample size.

Value

A pc_cv object containing:

• samples: Vector of sample sizes used

• results: Array of causality results

• parameters: List of analysis parameters

The results array structure depends on the bootstrap parameter:

• If bootstrap>1: A three-dimensional array where first dimension represents sample sizes, second dimension contains statistics (mean, quantiles, median), and third dimension represents causality types (positive, negative, dark)

• If bootstrap=1: A three-dimensional array where first dimension represents sample sizes, second dimension contains single values, and third dimension represents causality types (positive, negative, dark)

See Also

plot.pc_cv for visualizing cross-validation results print.pc_cv for printing cross-validation results summary.pc_cv for summarizing cross-validation results

Examples

data(climate_indices)
X <- climate_indices$AO
Y <- climate_indices$AAO

# Basic cross-validation
cv_result <- pcCrossValidation(
  X, Y, 
  E = 3, tau = 1,
  metric = "euclidean",
  h = 1,
  weighted = FALSE,
  numberset = c(100, 200, 300)
)

# Cross-validation with bootstrap
cv_result_boot <- pcCrossValidation(
  X, Y,
  E = 3, tau = 1,
  metric = "euclidean",
  h = 1,
  weighted = FALSE,
  numberset = c(100, 200, 300),
  random = TRUE,
  bootstrap = 100
)

Pattern Causality Effect Analysis

Description

Analyzes pattern causality matrices to compute and summarize the directional effects of different causality types (positive, negative, dark) between system components.

Usage

pcEffect(pcmatrix, verbose = FALSE)

Arguments

Details

Calculate Pattern Causality Effect Analysis

The function performs these key steps:

• Processes raw causality matrices

• Computes received and exerted influence for each component

• Calculates net causality effect (difference between received and exerted)

• Normalizes results to percentage scale

Value

An object of class "pc_effect" containing:

• positive: Data frame of positive causality effects

• negative: Data frame of negative causality effects

• dark: Data frame of dark causality effects

• items: Vector of component names

• summary: Summary statistics for each causality type

Related Packages

• vars: Vector autoregression for multivariate time series

• lmtest: Testing linear regression models

• causality: Causality testing and modeling

See Also

pcMatrix for generating causality matrices plot.pc_effect for visualizing causality effects

Examples

data(climate_indices)
dataset <- climate_indices[, -1]
pcmatrix <- pcMatrix(dataset, E = 3, tau = 1, 
                    metric = "euclidean", h = 1, 
                    weighted = TRUE)
effects <- pcEffect(pcmatrix)
print(effects)
plot(effects)

Calculate Full Details Pattern Causality Analysis

Description

Implements an advanced pattern causality algorithm to explore the causal relationships between two time series datasets. This function provides comprehensive analysis of causality patterns, including state space reconstruction, pattern identification, and causality strength evaluation.

Usage

pcFullDetails(
  X,
  Y,
  E,
  tau,
  h,
  weighted,
  metric = "euclidean",
  distance_fn = NULL,
  state_space_fn = NULL,
  relative = TRUE,
  verbose = FALSE
)

Arguments

Details

Calculate Full Details Pattern Causality Analysis

The function implements these key steps:

• State Space Reconstruction: Creates shadow attractors using embedding

• Pattern Analysis: Converts time series into signature and pattern spaces

• Nearest Neighbor Analysis: Identifies and analyzes local dynamics

• Causality Evaluation: Computes predicted and actual causality matrices

• Results Validation: Provides detailed diagnostics and quality metrics

Value

A pc_full_details object containing:

• backtest_time: Time points used for backtesting

• valid_time: Valid time points for analysis

• causality_real: Real causality spectrum

• causality_pred: Predicted causality spectrum

• state_spaces: State space reconstructions

• neighbors: Nearest neighbor information

• patterns: Pattern and signature information

• matrices: Causality matrices

• predictions: Predicted and actual values

• weighted: A logical indicating if weighted calculations were used

• E: Embedding dimension used for the analysis

Calculate Pattern Causality Using Lightweight Algorithm

Description

Implements a computationally efficient version of the Pattern Causality Model Mk. II for analyzing causal interactions between two time series. This function uses pattern and signature spaces to assess causality through reconstructed state spaces and hashed pattern analysis.

Usage

pcLightweight(
  X,
  Y,
  E,
  tau,
  h,
  weighted,
  metric = "euclidean",
  distance_fn = NULL,
  state_space_fn = NULL,
  relative = TRUE,
  verbose = FALSE
)

Arguments

Details

Calculate Pattern Causality Using Lightweight Algorithm

The function implements these key steps:

• State space reconstruction using embedding parameters

• Pattern and signature space transformation

• Nearest neighbor analysis in reconstructed spaces

• Causality strength calculation using prediction accuracy

• Classification of causality types (positive/negative/dark)

Value

An object of class "pc_fit" containing:

• total: Total causality strength (0-1)

• positive: Proportion of positive causality (0-1)

• negative: Proportion of negative causality (0-1)

• dark: Proportion of dark causality (0-1)

See Also

pcFullDetails for detailed analysis pcMatrix for analyzing multiple time series

Examples

data(climate_indices)
X <- climate_indices$AO
Y <- climate_indices$AAO
result <- pcLightweight(X, Y, E = 3, tau = 1, 
                       metric = "euclidean", h = 2, 
                       weighted = TRUE, verbose = FALSE)
print(result)
summary(result)
plot(result)

Pattern Causality Matrix Analysis

Description

Analyzes pattern causality relationships between multiple time series by computing pairwise causality measures and organizing them into matrices.

Usage

pcMatrix(
  dataset,
  E,
  tau,
  metric = "euclidean",
  h,
  weighted = TRUE,
  distance_fn = NULL,
  state_space_fn = NULL,
  relative = TRUE,
  verbose = FALSE,
  n_cores = 1
)

Arguments

Details

Compute Pattern Causality Matrix Analysis

The function performs these key steps:

• Validates input data and parameters

• Computes pairwise causality measures

• Organizes results into causality matrices

• Provides summary statistics for each causality type

Value

A pc_matrix object containing causality matrices

Related Packages

• vars: Vector autoregression analysis

• tseries: Time series analysis tools

• forecast: Time series forecasting methods

Pattern Causality Cross-Validation Object

Description

Creates a pattern causality cross-validation object containing results from repeated sampling analysis. This function constructs an object of class pc_cv to store the results of cross-validation analysis.

Usage

pc_cv(samples = NULL, results = NULL, parameters = NULL)

Arguments

Value

An object of class "pc_cv".

Pattern Causality Effect Object

Description

Creates a pattern causality effect object that contains information about the received and exerted influences for different causality types. This function constructs an object of class pc_effect to store the results of effect analysis.

Usage

pc_effect(positive = NULL, negative = NULL, dark = NULL, items = NULL)

Arguments

Value

An object of class "pc_effect".

Pattern Causality Matrix Object

Description

Creates a pattern causality matrix object. This function constructs an object of class pc_matrix containing the positive, negative, and dark causality matrices, along with item names.

Usage

pc_matrix(
  positive = NULL,
  negative = NULL,
  dark = NULL,
  items = NULL,
  verbose = TRUE
)

Arguments

Value

An object of class "pc_matrix".

Examples

data(climate_indices)
dataset <- climate_indices[, -1]
pc_matrix_obj <- pcMatrix(dataset, E = 3, tau = 1, 
  metric = "euclidean", h = 1, weighted = TRUE, 
  verbose = FALSE)
print(pc_matrix_obj)

Pattern Causality Parameter Optimization Results

Description

Creates an object containing parameter optimization results for pattern causality analysis

Usage

pc_params(accuracy_summary, computation_time, parameters)

Arguments

Details

Pattern Causality Parameter Optimization Results

Value

An object of class "pc_params"

Plot Pattern Causality Cross Validation Results

Description

Visualizes the pattern causality cross-validation results. This function generates a line plot showing the causality strengths for different sample sizes.

Usage

## S3 method for class 'pc_cv'
plot(x, fr = FALSE, separate = FALSE, ...)

Arguments

Value

Invisibly returns the input object.

Examples

data(climate_indices)
X <- climate_indices$AO
Y <- climate_indices$AAO
numbersets <- c(100, 150, 200)
cv_results <- pcCrossValidation(X, Y, 3, 2, "euclidean", 1, FALSE, numberset = numbersets)
plot(cv_results)

Plot Pattern Causality Effect

Description

Generates a plot to visualize the effects of positive, negative, or dark causality. Displays the influence exerted versus influence received for each item. This function generates a scatter plot showing the influence exerted versus influence received for each item, colored by the difference between exerted and received influence.

Usage

## S3 method for class 'pc_effect'
plot(
  x,
  status = "positive",
  add_label = TRUE,
  point_size = 3,
  label_size = 3,
  ...
)

Arguments

Value

Invisibly returns the ggplot object.

Examples

data(climate_indices)
dataset <- climate_indices[, -1]
pc_matrix_obj <- pcMatrix(dataset, E = 3, tau = 1, 
  metric = "euclidean", h = 1, weighted = TRUE, 
  verbose = FALSE)
effects <- pcEffect(pc_matrix_obj)
plot(effects, status = "positive")

Plot Pattern Causality Results

Description

Generates a combined plot of total causality and causality components for a pc_fit object. This function combines the visualizations from plot_total and plot_components into a single plot.

Usage

## S3 method for class 'pc_fit'
plot(x, ...)

Arguments

Value

NULL invisibly.

Plot Pattern Causality Matrix

Description

Creates a heatmap visualization of the pattern causality matrix for positive, negative, or dark causality relationships. This function generates a heatmap using ggplot2 to visualize the specified causality matrix.

Usage

## S3 method for class 'pc_matrix'
plot(
  x,
  status,
  width = 0.85,
  height = 0.75,
  radius = grid::unit(3, "pt"),
  alpha = 0.53,
  show_text = FALSE,
  show_legend_title = FALSE,
  ...
)

Arguments

Value

A ggplot object invisibly.

References

Stavroglou et al. (2020) doi:10.1073/pnas.1918269117

Examples

data(climate_indices)
dataset <- climate_indices[, -1]
pc_matrix_obj <- pcMatrix(dataset, E = 3, tau = 1, 
  metric = "euclidean", h = 1, weighted = TRUE, 
  verbose = FALSE)
plot(pc_matrix_obj, status = "positive")

Plot State Space Reconstruction

Description

Visualizes the state space reconstruction in 3D. This function generates a 3D scatter plot of the reconstructed state space.

Usage

## S3 method for class 'pc_state'
plot(x, style = 2, verbose = FALSE, ...)

Arguments

Value

Invisibly returns the input object.

Plot Pattern Causality Time Series

Description

Visualizes the positive, negative and dark causality components over time

Usage

plot_causality(x, type, ...)

Arguments

Value

Invisibly returns the ggplot object

Plot Pattern Causality Time Series

Description

Visualizes the positive, negative and dark causality components over time

Usage

## S3 method for class 'pc_full_details'
plot_causality(x, type, ...)

Arguments

Value

Invisibly returns the ggplot object

Plot Pattern Causality Components

Description

Visualizes the positive, negative, and dark causality components as a barplot. This function takes a pc_fit object and generates a barplot showing the strength of each causality component.

Usage

plot_components(x, ...)

Arguments

Value

NULL invisibly.

Examples

data(climate_indices)
X <- climate_indices$AO
Y <- climate_indices$AAO
pc_result <- pcLightweight(X, Y, E = 3, tau = 2, metric = "euclidean", h = 1, weighted = TRUE)
plot_components(pc_result)

Plot Causality Components

Description

Visualizes the positive, negative, and dark causality components as a barplot for a pc_fit object. This function generates a barplot showing the strength of each causality component.

Usage

## S3 method for class 'pc_fit'
plot_components(x, ...)

Arguments

Value

NULL.

Plot Total Pattern Causality

Description

Visualizes the total pattern causality strength as a barplot. This function takes a pc_fit object and generates a barplot showing the overall causality strength.

Usage

plot_total(x, ...)

Arguments

Value

NULL invisibly.

References

Stavroglou et al. (2020) doi:10.1073/pnas.1918269117

See Also

plot_components for visualizing individual causality components.

Examples

data(climate_indices)
X <- climate_indices$AO
Y <- climate_indices$AAO
pc_result <- pcLightweight(X, Y, E = 3, tau = 2, metric = "euclidean", h = 1, weighted = TRUE)
plot_total(pc_result)

Plot Total Causality

Description

Visualizes the total causality strength as a barplot for a pc_fit object. This function generates a barplot showing the total causality strength and its complement.

Usage

## S3 method for class 'pc_fit'
plot_total(x, ...)

Arguments

Value

NULL.

Print Method for Pattern Causality Accuracy Results

Description

Print Method for Pattern Causality Accuracy Results

Usage

## S3 method for class 'pc_accuracy'
print(x, verbose = FALSE, ...)

Arguments

Value

Invisibly returns the input object

Print Pattern Causality Cross Validation Results

Description

Prints the pattern causality cross-validation results. This function displays the parameters used for cross-validation, the sample sizes, and the summary statistics.

Usage

## S3 method for class 'pc_cv'
print(x, ...)

Arguments

Value

Invisibly returns the input object.

Examples

data(climate_indices)
X <- climate_indices$AO
Y <- climate_indices$AAO
numberset <- c(100, 150, 200)
cv_results <- pcCrossValidation(X, Y, 3, 2, "euclidean", 1, FALSE, numberset = numberset)
print(cv_results)

Print Pattern Causality Effect

Description

Prints the pattern causality effect analysis results. This function displays the received and exerted influences for each item for positive, negative, and dark causality types.

Usage

## S3 method for class 'pc_effect'
print(x, ...)

Arguments

Value

Invisibly returns the input object.

Examples

data(climate_indices)
dataset <- climate_indices[, -1]
pc_matrix_obj <- pcMatrix(dataset, E = 3, tau = 1, 
  metric = "euclidean", h = 1, weighted = TRUE, 
  verbose = FALSE)
effects <- pcEffect(pc_matrix_obj)
print(effects)

Print Pattern Causality Results

Description

Prints the pattern causality analysis results from a pc_fit object. This function displays the total, positive, negative, and dark causality strengths.

Usage

## S3 method for class 'pc_fit'
print(x, ...)

Arguments

Value

Invisibly returns the input object.

Print Pattern Causality Matrix

Description

Prints the pattern causality matrix object. This function displays the specified causality matrix (or all matrices) with a preview of the first 5 rows and columns.

Usage

## S3 method for class 'pc_matrix'
print(x, type = "all", ...)

Arguments

Value

Invisibly returns the input object.

Examples

data(climate_indices)
dataset <- climate_indices[, -1]
pc_matrix_obj <- pcMatrix(dataset, E = 3, tau = 1, 
  metric = "euclidean", h = 1, weighted = TRUE, 
  verbose = FALSE)
print(pc_matrix_obj, type = "positive")

Print Method for Pattern Causality Parameter Results

Description

Print Method for Pattern Causality Parameter Results

Usage

## S3 method for class 'pc_params'
print(x, verbose = FALSE, ...)

Arguments

Value

Invisibly returns the input object

Print State Space Reconstruction

Description

Prints the state space reconstruction results. This function displays the parameters used for state space reconstruction and a preview of the reconstructed points.

Usage

## S3 method for class 'pc_state'
print(x, ...)

Arguments

Value

Invisibly returns the input object.

Print Method for Pattern Causality Accuracy Summary

Description

Print Method for Pattern Causality Accuracy Summary

Usage

## S3 method for class 'summary.pc_accuracy'
print(x, ...)

Arguments

Value

Invisibly returns the input object

State Space Reconstruction

Description

Reconstructs the state space of a time series using delay embedding, creating a matrix where each row represents a point in the reconstructed space.

Usage

stateSpace(ts, E, tau, verbose = FALSE)

Arguments

Details

State Space Reconstruction Analysis

The function implements Takens' embedding theorem to reconstruct state space:

• Creates delay vectors using specified embedding dimension (E)

• Applies time delay (tau) between consecutive elements

• Handles boundary conditions and missing values

Value

An object of class "pc_state" containing:

• matrix: The reconstructed state space matrix

• parameters: List of reconstruction parameters

• original: Original time series data

Related Packages

• nonlinearTseries: Nonlinear time series analysis

• tseriesChaos: Chaos theory analysis tools

• fractal: Fractal analysis methods

Examples

ts <- c(1:100)
result <- stateSpace(ts, E = 3, tau = 2)
plot(result)

State Space Reconstruction Interface

Description

A generic interface for reconstructing state spaces from time series data using either built-in or custom methods.

Usage

stateSpaceMethod(x, E, tau, ...)

## Default S3 method:
stateSpaceMethod(x, E, tau, ...)

## S3 method for class 'custom'
stateSpaceMethod(x, E, tau, method, ...)

Arguments

Details

Generic Interface for State Space Reconstruction

Value

A list containing the reconstructed state space components

Methods (by class)

• stateSpaceMethod(default): Default state space reconstruction

• stateSpaceMethod(custom): Custom state space reconstruction

Examples

## Not run: 
# Using default method
x <- rnorm(100)
s1 <- stateSpaceMethod(x, E=3, tau=2)

# Using custom method
custom_space <- function(x, E, tau) {
  list(matrix=embed(x, E))
}
s2 <- stateSpaceMethod(x, E=3, tau=2, method=custom_space)

## End(Not run)

Summary Method for Pattern Causality Accuracy Results

Description

Summary Method for Pattern Causality Accuracy Results

Usage

## S3 method for class 'pc_accuracy'
summary(object, ...)

Arguments

Value

A summary object for pc_accuracy

Summary of Pattern Causality Cross Validation Results

Description

Provides a summary of the pattern causality cross-validation results. This function calculates and displays summary statistics for the cross-validation results, including sample statistics, causality statistics, and convergence.

Usage

## S3 method for class 'pc_cv'
summary(object, ...)

Arguments

Value

Invisibly returns the input object.

Examples

data(climate_indices)
X <- climate_indices$AO
Y <- climate_indices$AAO
numberset <- c(100, 150, 200)
cv_results <- pcCrossValidation(X, Y, 3, 2, "euclidean", 1, FALSE, numberset = numberset)
summary(cv_results)

Summarize Pattern Causality Effect

Description

Provides a summary of the pattern causality effect analysis results. This function displays the summary statistics for the effects, including the number of components and the strongest effects.

Usage

## S3 method for class 'pc_effect'
summary(object, ...)

Arguments

Value

Invisibly returns the input object.

Examples

data(climate_indices)
dataset <- climate_indices[, -1]
pc_matrix_obj <- pcMatrix(dataset, E = 3, tau = 1, 
  metric = "euclidean", h = 1, weighted = TRUE, 
  verbose = FALSE)
effects <- pcEffect(pc_matrix_obj)
summary(effects)

Summarize Pattern Causality Results

Description

Provides a summary of the pattern causality analysis results from a pc_fit object. This function displays a table of causality strengths for total, positive, negative, and dark components.

Usage

## S3 method for class 'pc_fit'
summary(object, ...)

Arguments

Value

Invisibly returns the input object.

Summarize Pattern Causality Matrix

Description

Provides a summary of the pattern causality matrix object. This function calculates and displays descriptive statistics (mean, SD, min, max) for each causality matrix (positive, negative, dark).

Usage

## S3 method for class 'pc_matrix'
summary(object, ...)

Arguments

Value

Invisibly returns the input object.

Examples

data(climate_indices)
dataset <- climate_indices[, -1]
pc_matrix_obj <- pcMatrix(dataset, E = 3, tau = 1, 
  metric = "euclidean", h = 1, weighted = TRUE, 
  verbose = FALSE)
summary(pc_matrix_obj)

Summary Method for Pattern Causality Parameter Results

Description

Summary Method for Pattern Causality Parameter Results

Usage

## S3 method for class 'pc_params'
summary(object, ...)

Arguments

Value

A summary object for pc_params

Summarize State Space Reconstruction

Description

Provides a summary of the state space reconstruction results. This function displays the dimensions, number of points, parameters, summary statistics for each dimension, and the number of missing values.

Usage

## S3 method for class 'pc_state'
summary(object, ...)

Arguments

Value

Invisibly returns the input object.

Validate Custom Function Output for Pattern Causality Analysis

Description

Validates the Output Format from Custom Distance and State Space Functions to ensure compatibility with the package's internal processing.

Usage

validate_custom_fn_output(output, fn_name)

Arguments

Details

Validate Custom Function Output

Value

Nothing. Throws an error if validation fails.

Examples

# Example 1: Validating custom distance function output
custom_dist <- function(x) {
  # Create distance matrix
  dist_mat <- as.matrix(dist(x))
  # Validate output
  validate_custom_fn_output(dist_mat, "distance_fn")
  return(dist_mat)
}

# Example 2: Validating custom state space function output
custom_state_space <- function(x, E, tau) {
  # Create state space matrix
  n <- length(x) - (E-1)*tau
  state_mat <- matrix(nrow = n, ncol = E)
  for(i in 1:E) {
    state_mat[,i] <- x[1:n + (i-1)*tau]
  }
  # Create output list
  result <- list(matrix = state_mat, 
                parameters = list(E = E, tau = tau))
  # Validate output
  validate_custom_fn_output(result, "state_space_fn")
  return(result)
}

# Using the custom functions
x <- sin(seq(0, 4*pi, length.out = 100))
dist_result <- custom_dist(x)
space_result <- custom_state_space(x, E = 3, tau = 2)