Package: lingamr 0.1.2

Sample Data | generate_lingam_sample_6() | Causal Discovery | Causal Order | Estimated Adjacency Matrix | Drawing the Causal Graph | Comparing the Estimated and True Structures | Static Plotting with ggplot2 | Total Causal Effect | Comparison with Multiple Regression Coefficients | Inference with Prior Knowledge | Format of the Prior Knowledge Matrix | Usage Example | Specifying by Index | Specifying by Variable Name | Running Direct LiNGAM with Prior Knowledge | Choosing a Regression Method (reg_method) | Comparison of the Four Methods | Choosing lambda (common to LASSO / Adaptive LASSO) | Independence between Error Variables | The Non-Gaussianity Assumption | Non-Gaussian Errors (Uniform Distribution) -- When It Works | Gaussian Errors -- When It Fails | Testing the Normality of Residuals | Model Summary | Bootstrap Direct LiNGAM | Inspecting the Bootstrap Results | Adjacency Matrix of Mean Causal Effects | Matrix of Path Occurrence Frequencies | Mean Total Effects | Stability of the Causal Order | Integration with broom (tidy / glance) | A Larger Dataset (10 Variables) | Comparing ICA-LiNGAM and Direct LiNGAM | Running Both Algorithms | Comparing the Estimated Coefficients | Comparing the DAG Structures | When There Are Many Variables: The Scalability Wall | Generating the Data | Comparing Execution Times | Checking Estimation Accuracy (p = 10) | High-Dimensional Direct LiNGAM | A Case Where DirectLiNGAM Struggles: The Measurement Error Paradox | VAR-LiNGAM: Causal Discovery in Time Series | Fitting VAR-LiNGAM | Lag Order Selection | Stationarity Check | Residual Diagnostics | Total Causal Effects | Bootstrap | LiNGAM for Mixed Data (LiM) | Multi-Group Direct LiNGAM | Causal Discovery with Missing Data | Latent Confounders: Bottom-Up ParceLiNGAM | Latent Confounders: RCD | Evaluating Model Fit | When LiNGAM Cannot Be Used | A Checklist to Verify in Advance