Man pages for lgspline
Lagrangian Multiplier Smoothing Splines for Smooth Function Estimation
| AGAmult_wrapper | Efficient Matrix Multiplication for... |
| approx_grad | Finite-difference Gradient Computer |
| armaInv | Matrix Inversion using Armadillo |
| blockfit_solve | Backfitting Solver for Blockfit Models |
| coef.lgspline | Extract Coefficients from a Fitted lgspline |
| coef.wald_lgspline | Extract Coefficients from a wald_lgspline Object |
| collapse_block_diagonal | Collapse Matrix List into a Single Dense Block-Layout Matrix |
| compute_dG_dlambda | Compute Derivative of Inverse-Information Matrix G with... |
| compute_dGhalf | Compute Eigen-Based Square-Root Factors for... |
| compute_dG_u_dlambda_xy | Derivative of Constrained Penalized Coefficients with Respect... |
| compute_G_eigen | Compute Eigenvalues and Related Matrices for G |
| compute_GhalfXy_temp_wrapper | Compute Component... |
| compute_gram_block_diagonal | Compute Gram Matrix for Block Diagonal Structure |
| compute_Lambda | Construct Smoothing Spline Penalty Matrix |
| compute_trace_H | Effective degrees of freedom via trace of the hat matrix |
| compute_trace_UGXX_wrapper | Calculate Trace of Matrix Product... |
| confint.lgspline | Confidence Intervals for lgspline Coefficients |
| confint.wald_lgspline | Extract Confidence Intervals from a wald_lgspline Object |
| cox_dispersion_function | Cox PH Dispersion Function |
| cox_family | Cox Proportional Hazards Family for lgspline |
| cox_glm_weight_function | Cox PH GLM Weight Function |
| cox_helpers | Cox Proportional Hazards Helpers for lgspline |
| cox_qp_score_function | Cox PH Score Function for Quadratic Programming and Blockfit |
| cox_schur_correction | Cox PH Schur Correction |
| create_block_diagonal | Create Block Diagonal Matrix |
| create_onehot | Create One-Hot Encoded Matrix |
| damped_newton_r | Damped Newton-Raphson Parameter Optimization |
| Details | Lagrangian Multiplier Smoothing Splines: Mathematical Details |
| dot-active_set_refine | Partition-Wise Active-Set Refinement for Inequality... |
| dot-active_set_refine_full | Full-System Active-Set Refinement for Correlated Paths |
| dot-active_set_refine_woodbury | Partition-Wise Active-Set Refinement for Woodbury GEE Paths |
| dot-bf_assemble_qp_info | Assemble qp_info from a solve.QP Solution |
| dot-bf_case_gauss_gee | Gaussian Identity + GEE: Closed-Form Solve (Case a) |
| dot-bf_case_gauss_no_corr | Gaussian Identity Backfitting Without Correlation (Case b) |
| dot-bf_case_glm_gee | GLM + GEE Two-Stage Estimation (Case c) |
| dot-bf_case_glm_no_corr | GLM Without GEE: Damped Newton-Raphson + Backfitting (Case d) |
| dot-bf_constrained_flat_update | Constrained Flat Update Given Current Spline Solution |
| dot-bf_deviance | Compute Deviance for Non-GEE Models |
| dot-bf_gee_deviance | Compute Whitened Deviance for GEE Convergence Monitoring |
| dot-bf_get_mu_eta | Numerical Derivative of the Inverse Link Function |
| dot-bf_inner_weighted | GLM Backfitting Inner Loop |
| dot-bf_lagrangian_project | Lagrangian Projection for the Spline-Only Subproblem |
| dot-bf_make_Lambda_block | Construct Full Block-Diagonal Penalty Matrix |
| dot-bf_split_components | Split Design and Penalty into Spline and Flat Components |
| dot-bf_sqp_loop | Damped SQP Refinement on the Full (Optionally Whitened)... |
| dot-build_deriv_qp | Build Derivative QP Constraints in Full P-Dimensional Space |
| dot-build_lambda_block | Build Block-Diagonal Penalty Matrix |
| dot-build_tuning_env | Build Tuning Environment |
| dot-check_kkt_full | Check KKT Conditions for Full-System Active-Set Refinement |
| dot-check_kkt_partitionwise | Check KKT Conditions for Partition-Wise Active-Set Method |
| dot-check_kkt_partitionwise_woodbury | Check KKT Conditions for Woodbury Partition-Wise Active-Set... |
| dot-compute_dhat_diag | Compute the Derivative of the Tuning Hat Matrix Diagonal |
| dot-compute_dist_block | Compute Euclidean distance matrix for a cluster block |
| dot-compute_gcvu | Evaluate Modified GCV_u Criterion at a Given Penalty... |
| dot-compute_gcvu_gradient | Compute Closed-Form Gradient of GCV_u Criterion |
| dot-compute_hat_diag | Compute the Diagonal of the Tuning Hat Matrix |
| dot-compute_loocv | Evaluate Exact Leave-One-Out Criterion at a Given Penalty... |
| dot-compute_loocv_gradient | Compute Closed-Form Gradient of Exact Leave-One-Out Criterion |
| dot-compute_meta_penalty | Compute Regularization (Meta) Penalty on Penalty Parameters |
| dot-compute_meta_penalty_gradient | Compute Gradient of Regularization (Meta) Penalty |
| dot-compute_score_V_partitioned | Compute Partitioned GEE Score Vector |
| dot-compute_tuning_delta | Compute Pseudocount Delta for Link Function Stabilization |
| dot-compute_tuning_predictions | Compute Predictions During Penalty Tuning |
| dot-compute_tuning_residuals | Compute Residuals During Penalty Tuning |
| dot-compute_Xy_V | Compute Partitioned GLS Cross-Products |
| dot-constraint_col_scales | Compute Stable Column Rescaling for Constraint Matrices |
| dot-damped_bfgs | Damped BFGS Optimizer for Penalty Tuning |
| dot-detect_qp_global | Detect Whether Inequality Constraints Need the Global Bridge |
| dot-extract_G_diagonal | Extract Per-Partition Diagonal Blocks from a Full Matrix |
| dot-fit_coefficients | Fit Coefficients During Penalty Tuning: blockfit_solve or... |
| dot-fit_get_B | Call get_B During Penalty Tuning |
| dot-get_B_gaussian_nocorr | Path 2: Gaussian Identity Link, No Correlation |
| dot-get_B_gee_gaussian | Path 1a: Gaussian Identity + GEE (Closed-Form Full-System... |
| dot-get_B_gee_glm | Path 1b: Non-Gaussian GEE (Damped SQP with Full Whitened... |
| dot-get_B_gee_glm_woodbury | Path 1b-Woodbury: Non-Gaussian GEE with Woodbury Acceleration |
| dot-get_B_gee_woodbury | Path 1a-Woodbury: Gaussian GEE with Woodbury Acceleration |
| dot-get_B_glm_nocorr | Path 3: Non-Gaussian GLM, No Correlation |
| dot-lagrangian_project | Lagrangian Projection via OLS Reformulation |
| dot-lagrangian_project_full | Full-System Lagrangian Projection |
| dot-lagrangian_project_woodbury | Woodbury-Corrected Lagrangian Projection |
| dot-qp_refine | Quadratic Programming Refinement for Inequality Constraints |
| dot-rank_dists | Compute rank-based distance matrix for AR(1) structures |
| dot-recompute_G_at_estimate | Recompute G at Current Coefficient Estimates |
| dot-solver_active_set | Generic Outer Active-Set Loop Around Equality-Constrained... |
| dot-solver_assemble_qp_info | Assemble qp_info from a solve.QP Solution |
| dot-solver_build_active_qp_info | Build qp_info for an Active-Set Solve |
| dot-solver_build_augmented_constraints | Build Augmented Equality System for an Active-Set Iteration |
| dot-solver_build_constraint_rhs | Build Equality Right-Hand Side for Constraint Matrix |
| dot-solver_build_lambda_block | Build Block-Diagonal Penalty Matrix |
| dot-solver_check_primal_feasibility | Check Primal Feasibility of Inactive Inequality Constraints |
| dot-solver_detect_qp_global | Detect Whether Inequality Constraints Need the Global Bridge |
| dot-solver_recompute_G_at_estimate | Recompute G, Ghalf, and GhalfInv at a Supplied Coefficient... |
| dot-solver_recover_active_multipliers | Recover Active-Constraint Multipliers from Transformed OLS |
| dot-solver_resolve_active_set | Re-Solve an Equality Subproblem for a Fixed Active Set |
| dot-solver_stabilize_working_weights | Stabilize GLM Working Weights Before Linear Algebra |
| dot-try_woodbury_active_set | Attempt Woodbury Partition-Wise Active-Set Refinement |
| dot-tune_grid_search | Grid Search Initialization for Penalty Tuning |
| dot-woodbury_decompose_V | Woodbury Decomposition of Correlation Structure |
| dot-woodbury_halfsqrt_components | Compute Woodbury Half-Square-Root Components |
| dot-woodbury_redecompose_weighted | Per-Iteration Weighted Woodbury Redecomposition |
| dot-woodbury_transform_constraint_matrix | Form the Woodbury-Corrected Constraint Design Matrix |
| dot-woodbury_transform_constraint_matrix_transpose | Form the Transpose Woodbury-Corrected Constraint Matrix |
| efficient_bfgs | BFGS Implementation for REML Parameter Estimation |
| efficient_matrix_mult | Efficient Matrix Multiplication |
| equation | Print Closed-Form Fitted Equation from lgspline Model |
| expgrid | Generate Grid Indices Without expand.grid() |
| find_extremum | Find the Extremum of a Fitted lgspline |
| find_neighbors | Find Neighboring Cluster Partitions Using Midpoint Distance... |
| GAmult_wrapper | Efficient Matrix Multiplication of G and A Matrices |
| generate_posterior | Generate Posterior Samples from a Fitted lgspline |
| generate_posterior_correlation | Generate Posterior Samples Propagating Correlation Parameter... |
| get_2ndDerivPenalty | Compute Integrated Squared Second Derivative Penalty Matrix |
| get_2ndDerivPenalty_wrapper | Wrapper for Integrated Second-Derivative Penalty Computation |
| get_B | Compute Constrained GLM Coefficient Estimates via Lagrangian... |
| get_B_verification_examples | Verification Examples for get_B |
| get_centers | Get Centers for Partitioning |
| get_interaction_patterns | Generate Interaction Variable Patterns |
| get_polynomial_expansions | Generate Design Matrix with Polynomial and Interaction Terms |
| get_U | Efficiently Construct U Matrix |
| gramMatrix | Compute Gram Matrix |
| grapes-times-times-grapes | Efficient Matrix Multiplication Operator |
| info_cox | Compute Cox Observed Information Matrix |
| info_negbin | Compute Negative Binomial Observed Information Matrix |
| integrate | Generic for Numerical Integration |
| integrate.lgspline | Definite Integral of a Fitted lgspline |
| invert | Matrix Inversion with Fallback Methods |
| is_binary | Test if Vector is Binary |
| knot_expand_list | Expand Matrix into Partition Lists Based on Knot Boundaries |
| leave_one_out | Compute Leave-One-Out Cross-Validated Predictions for... |
| lgspline | Fit Lagrangian Multiplier Smoothing Splines |
| lgspline_cox | Fit Cox Proportional Hazards Model via lgspline |
| lgspline.fit | Low-Level Fitting for Lagrangian Smoothing Splines |
| lgspline-methods | lgspline: S3 Methods |
| lgspline_negbin | Fit Negative Binomial Model via lgspline |
| lgspline-package | Lagrangian Multiplier Smoothing Splines |
| lgspline_weibull | Fit Weibull Accelerated Failure Time Model via lgspline |
| loglik_cox | Compute Cox Partial Log-Likelihood |
| logLik.lgspline | Extract Log-Likelihood from a Fitted lgspline |
| loglik_negbin | Compute Negative Binomial Log-Likelihood |
| loglik_weibull | Compute Log-Likelihood for Weibull Accelerated Failure Time... |
| make_constraint_matrix | Create Smoothing Spline Constraint Matrix |
| make_derivative_matrix | Compute First and Second Derivative Matrices |
| make_partitions | Create Data Partitions Using Clustering |
| matinvsqrt | Calculate Matrix Inverse Square Root for Symmetric Matrices |
| matmult_block_diagonal | Multiply Block Diagonal Matrices in Parallel |
| matmult_U | Left-Multiply a List of Block-Diagonal Matrices by U |
| matsqrt | Calculate Matrix Square Root for Symmetric Matrices |
| negbin_dispersion_function | NB Dispersion Function |
| negbin_family | Negative Binomial Family for lgspline |
| negbin_glm_weight_function | NB GLM Weight Function |
| negbin_helpers | Negative Binomial Regression Helpers for lgspline |
| negbin_qp_score_function | NB Score Function for Quadratic Programming and Blockfit |
| negbin_schur_correction | NB Schur Correction |
| negbin_theta | Estimate Negative Binomial Shape Parameter |
| nr_iterate | Compute Newton-Raphson Parameter Update with Numerical... |
| plot.lgspline | Plot Method for lgspline Objects |
| plot.wald_lgspline | Plot Method for wald_lgspline Objects |
| predict.lgspline | Predict Method for lgspline Objects |
| print.lgspline | Print Method for lgspline Objects |
| print.summary.lgspline | Print Method for lgspline Summaries |
| print.wald_lgspline | Print Method for wald_lgspline Objects |
| prior_loglik | Log-Prior Distribution Evaluation for lgspline Models |
| process_input | Process Input Arguments for lgspline |
| process_qp | Prepare Quadratic Programming Constraints for lgspline |
| reml_grad_from_dV | Evaluate the REML gradient with respect to a single... |
| safe_replace_var | Safely Replace Variable Names in Printed Terms |
| score_cox | Compute Cox Partial Log-Likelihood Score Vector |
| score_negbin | Compute Negative Binomial Score Vector |
| softplus | Compute softplus transform |
| std | Standardize Vector to Z-Scores |
| summary.lgspline | Summary Method for lgspline Objects |
| summary.wald_lgspline | Summary Method for wald_lgspline Objects |
| take_derivative | Calculate Derivatives of Polynomial Terms |
| take_interaction_2ndderivative | Calculate Second Derivatives of Interaction Terms |
| tune_Lambda | Tune Smoothing and Ridge Penalties via Leave-One-Out or... |
| unconstrained_fit_cox | Unconstrained Cox PH Estimation for lgspline |
| unconstrained_fit_default | Unconstrained Generalized Linear Model Estimation |
| unconstrained_fit_negbin | Unconstrained NB Estimation for lgspline |
| unconstrained_fit_weibull | Unconstrained Weibull Accelerated Failure Time Model... |
| vectorproduct_block_diagonal | Vector-Matrix Multiplication for Block Diagonal Matrices |
| wald_univariate | Univariate Wald Tests and Confidence Intervals for lgspline... |
| weibull_dispersion_function | Estimate Weibull Dispersion for Accelerated Failure Time... |
| weibull_family | Weibull Family for Survival Model Specification |
| weibull_glm_weight_function | Weibull GLM Weight Function for Constructing Information... |
| weibull_qp_score_function | Compute Gradient of Log-Likelihood of Weibull Accelerated... |
| weibull_scale | Estimate Scale for Weibull Accelerated Failure Time Model |
| weibull_schur_correction | Correction for the Variance-Covariance Matrix for Uncertainty... |
