Package index • FunQ

Core Methodologies

Functions implementing main quantile PCA algorithms

fqpca(): FQPCA (Functional Quantile Principal Component Analysis)
mfqpca(): MFQPCA (Multilevel Functional Quantile Principal Component Analysis)

Parameter Cross-Validation

Functions to perform k-fold cross-validation on hyperparameters

fqpca_cv_df(): Cross validation of degrees of freedom
fqpca_cv_lambda(): Cross validation of lambda.ridge
mfqpca_cv_df(): Cross validation of degrees of freedom

S3 Methods & Plotting

Methods to plot, fit, and predict from fitted model objects

fitted(<fqpca_object>): Fit Yhat
fitted(<mfqpca_object>): Fit Yhat
plot(<fqpca_object>): Plot fqpca loading functions with ggplot2
plot(<mfqpca_object>): Plot fqpca loading functions with ggplot2
predict(<fqpca_object>): Predict fqpca scores
predict(<mfqpca_object>): Predict mfqpca scores

Utilities & Helper Metrics

Internal calculations and error metrics

create_folds(): Split a given matrix Y into k-folds
train_test_split(): Split a given matrix Y into train / test
quantile_error(): Quantile error computation
proportion_under_quantile(): Proportion of points under quantile

Internal Utilities

Internal helper functions not exported by the package

build_tensor_matrix(): Build Tensor Design Matrix
build_tensor_matrix_sparse(): Build Sparse Tensor Design Matrix
check_flag(): Check that a parameter is a single Boolean
check_fqpca_params(): Check fqpca input parameters
check_iters(): Check that a maximum-iterations parameter is an integer >= 2
check_loss(): Quantile check (pinball) loss
check_nonnegative_number(): Check that a parameter is a non-negative number
check_positive_float(): Check that a parameter is a positive float
check_positive_integer(): Check that a parameter is a positive integer
check_quantile_value(): Check that the quantile value lies in (0, 1)
check_regressors(): Checks input regressors
check_seed(): Check that the seed is a positive integer or NULL
check_splines_method(): Check that the splines method is supported
coerce_functional_input(): Coerce functional input data
compute_loadings(): Compute loadings (aka principal components)
compute_objective_value(): Compute objective value function.
compute_scores(): Compute scores
cv_framework(): Generic cross-validation framework
explained_variance_ratio(): Explained variance ratio of the scores
extract_coefficients(): Extract the coefficient vector from a quantile regression model object
fit_tensor_quantile_regression(): Fit the tensor-design quantile regression
kfold_cv_points(): Split a given matrix Y into k-folds
kfold_cv_rows(): Split a given matrix Y into k-folds
mfqpca_check_params(): Check mfqpca input parameters
mfqpca_compute_scores_between(): Compute scores between
reconstruct_scores_loadings(): Truncated reconstruction from scores and loadings
rotate_factors(): Rotation of loadings and scores
select_npc(): Select the number of components
train_test_split_points(): Split points of a given matrix Y into train / test
train_test_split_rows(): Split rows of a given matrix Y into train / test