Package index • risdr

Package overview

risdr risdr-package: risdr: Regularised and Information-Theoretic Sufficient Dimension Reduction

Model fitting and methods

fit_risdr(): Fit regularised and information-theoretic SDR model
compute_dr(): Compute Directional Regression
compute_information_criteria(): Compute model selection criteria
compute_phd(): Compute Principal Hessian Directions
compute_save(): Compute Sliced Average Variance Estimation
compute_scores(): Compute SDR scores for new data
compute_sdr(): General SDR kernel dispatcher
compute_sir(): Compute Sliced Inverse Regression
extract_loadings(): Extract SDR loadings
predict(<risdr>): Predict method for risdr objects
predict_downstream_lm(): Predict from downstream SDR regression model
prediction_correlation(): Prediction correlation
summary(<risdr>): Summarise risdr object
print(<risdr>): Print risdr object
print(<risdr_realdata>): Print real-data RISDR workflow
print(<summary.risdr>): Print summary of risdr object

Covariance estimation and stabilisation

cov_complexity_C1(): Covariance complexity C1
cov_complexity_C1F(): Scale-invariant covariance complexity C1F
cov_condition_number(): Covariance condition number
cov_diagnostics(): Covariance eigenvalue summary
cov_effective_rank(): Effective covariance rank
cov_lw(): Ledoit-Wolf type covariance estimator
cov_mec(): Maximum Entropy Covariance estimator
cov_oas(): Oracle Approximating Shrinkage covariance estimator
cov_ridge(): Ridge-type covariance estimator
cov_sample(): Sample covariance matrix
estimate_cov(): General covariance estimator dispatcher
stabilize_cov(): General covariance stabilisation dispatcher
stabilize_eigenfloor(): Eigenvalue floor stabilisation
stabilize_nearest_pd(): Nearest positive definite stabilisation
stabilize_ridge(): Ridge stabilisation of covariance matrix

Structural dimension selection

select_dimension(): Select structural dimension
choose_dimension(): Choose dimension from criteria table
criterion_weights(): Information criterion weights
select_dimension_cv(): Cross-validation dimension selection for SDR
select_dimension_cv_icomp(): Complexity-aware cross-validation for structural dimension selection
select_dimension_ladle(): Ladle-type structural dimension diagnostic
make_cv_folds(): Create cross-validation folds
plot_cv_dimension(): Plot cross-validation dimension selection result
plot_dimension_selection(): Plot dimension selection criteria
plot_ladle(): Plot ladle dimension diagnostic
plot_loadings(): Plot SDR loadings
plot_prediction(): Prediction plot
plot_residuals(): Residual plot
plot_scree(): Scree plot of SDR eigenvalues
plot_sufficient(): Sufficient summary plot
plot_sufficient2d(): Two-direction sufficient summary plot

Prediction and assessment

fit_downstream_lm(): Fit downstream regression model on SDR scores
predict_downstream_lm(): Predict from downstream SDR regression model
evaluate_prediction(): Evaluate predictions
evaluate_prediction_cv(): Repeated cross-validation for predictive assessment
rmse(): Root mean squared error
mae(): Mean absolute error
mape(): Mean absolute percentage error
r_squared(): Coefficient of determination
adjusted_r_squared(): Adjusted coefficient of determination
prediction_correlation(): Prediction correlation

Slicing and simulation

make_slices(): Create response slices
slice_covariances(): Compute slice covariance matrices
slice_means(): Compute slice means
slice_proportions(): Compute slice proportions
slice_summary(): Summarise response slices
simulate_risdr_data(): Simulate SDR data
run_one_simulation(): Run one SDR simulation replication
run_risdr_simulation(): Run SDR simulation study
summarise_simulation(): Summarise simulation results
ar1_covariance(): AR(1) covariance matrix
projection_matrix(): Projection matrix
subspace_distance(): Subspace distance

Real-data helpers

filter_low_variance(): Filter low-variance predictors
prepare_survival_response(): Prepare survival response
fit_risdr_realdata(): Fit RISDR to real high-dimensional data