All functions

addMissingSpecColumns()

Add missing columns for specification

all_match() all_match()

check if all values of two vectors match

avg_spacing()

Computes means for various slices of a regression_spec by betas product.

bin_along_range()

Bin a variable along a range

bootstrapRows() bootstrapRows()

Function that fully bootstraps rows

capture_output()

Capture print output

check_algorithm()

Check if algorithm exists

coef(<qs>)

Method for getting coefficients from fitted qs model

collapse_correctly()

helper function, collapse using correct method

columns_match_vector() columns_match_vector()

Check which rows of a matrix match a whole vector

cub_root()

Find cube root of the form ax^3 + bx^2 + cx + d=0

cub_root_deriv()

Derivative of a cubic root

cub_root_select()

Find cube root of the form ax^3 + bx^2 + cx + d=0

cub_root_select_rconics()

Calculate cubic roots using the Rconics package

defcombine()

helper function, borrowed from foreach

denseMatrixToSparse()

Convert matrix to a SparseM csr matrix

distributional_effects()

Calculate distributional effects

do_matched_call()

run function only with arguments that match the arguments for f

.onAttach()

Set bootstrap cores on load + welcome messages

ensureSpecFullRank()

Ensure that a regression specification is full rank

eval_CDF()

Evaluate CDF given quantiles and residuals

eval_PDF()

Evaluate PDF given quantiles and residuals

eval_Quantiles()

Evaluate CDF given quantiles and residuals

eval_density_R()

A function to evaluate the quantile or density of given data based on normal distribution

findRedundantCols()

Returns the indices of the columns to remove to construct a full rank matrix

fitQuantileRegression()

Estimate a single quantile regression

fit_approx_quantile_model()

Compute quantile regression via accelerated gradient descent using Huber approximation, warm start based on data subset

fit_lasso()

Fit a quantile regression w/ a lasso penalty

getColNums()

Get column numbers given starting values and regression specification

getCores() setCores()

Get user defined cores

getRank()

Computes the rank of a matrix. Outputs consistent results across various matrix types.

getRows() getRows()

Function which gets resampled rows

getWeights() getWeights()

Function that returns the correct weights for weighted bootstrap

get_intercept()

Finds which column of X has the intercept term

get_marginal_effects()

Calculates the marginal effects of an N x p matrix (wide-format) of qreg coefficients

get_strata() get_strata()

Get clusters for subsampling given a formula

get_underlying()

Get the data inside the s4 slot of this sparse matrix class

inv()

Inverse of a matrix, but catches the error

lasso_cv_search()

Search for optimal lambda via cross-validation

leaveRows() leaveRows()

Function that doesn't reorder rows (for weighted bootstrap)

makePlan()

Make a plan for the future parallel backend

make_se_mat()

Make matrix into a "standard error" matrix

map_parallel()

Map a function along a list in parallel

map_rows_parallel()

Map a function along rows of a matrix or data.frame

marginal_effects()

Get all marginal effects of variables in the fit

me()

Get marginal effects at a set of levels for the covariates

me_by_variable()

Get marginal effects at a set of levels for the covariates

make_penalized_blots() mice.impute.qs()

Imputation function to be used with the mice package

na_if_null()

return NA if argument is null

pad_strings()

Pad vector of strings based on longest length

plot(<distributional_effects>)

Visualize distributional effects

predict(<qs>)

Predict quantiles given fitted spacings model

print(<qs>)

Print qs summary

print(<qs_summary>)

Print qs summary

printWarnings()

Generate warning messages for regression model

q_spline_R()

Computes the tail parameters and the second derivatives given quantiles

qs()

Compute quantile regressions via quantile spacings

qs_control()

Control quantreg_spacing parameters

quad_form()

Quadratic form of the cubic polynomial

quantreg_spacing()

Computes coefficients for the quantile regression spacing method.

randomly_assign()

Randomly assign fold ids

regressResiduals()

Runs quantile regression on residuals of the model (calculates spaces around jstar quantile)

reorder_coefficients()

Reorder coefficients in case intercept wasn't the first term

repMat()

For copying matrices as in Matlab (works for sparse matrices)

rescale_coefficients()

Rescale coefficients estimated on scaled data to match unscaled data

rho()

Function for calculating Pseudo-R^2 values Evaluates the check objective function (possibly weighted) for QREG

round_if()

Round if x is numeric, otherwise don't

rq.fit.agd()

Quantile Regression approximated w/ huber loss

rq.fit.br()

Version that complies with more general requirements

rq.fit.lasso()

Quantile Regression w/ Lasso Penalty

rq.fit.post_lasso()

Quantile Regression w/ Lasso Penalty

rq.fit.sfn()

Version that complies with more general requirements

rq.fit.sfn_start_val()

Sparse Regression Quantile Fitting with Weights

scale_for_lasso()

Scale matrix for lasso regression

se_control()

Control standard_errors parameters

seq_quant()

For sequencing along probabilities

spDiag()

Create sparse diagonal matrix with vector x on diagonal

spSums()

Return column sums of matrix

spacings_to_quantiles()

Compute quantiles given parameter coefficients and data

splint_R()

A function to conduct the interpolation given data and fitted quantiles

resample_qs() weighted_bootstrap() bootstrap() subsample() standard_errors() resample_qs() weighted_bootstrap() bootstrap() subsample() standard_errors()

Computes standard errors for the quantile regression spacing method using subsampling.

subsampleRows() subsampleRows()

Function that fully bootstraps rows

summary(<qs>)

creates a table of summary output for a qs object