get_suff_stats_kernel: Get loop-invariant quantities for empirical MCR binary search...

In aaronjfisher/mcr: Computes model class reliance, i.e., the most and least that models from a class may rely on variables of interest

Description

Creates 'nrep_sample' copies of the sample, each of which permutes the covariate of interest in a different way. (Primarily intended for internal use.)

Usage

get_suff_stats_kernel(y, X, dat_ref, kern_fun, p1 = NULL,
  reg_threshold = Inf, tol = 10^-10, nrep_sample = 2,
  verbose = TRUE, warn_psd = TRUE, warn_duplicate = TRUE,
  warn_dropped = TRUE)

Arguments

y	scalar outcome vector
X	matrix of covariates
dat_ref	reference matrix of covariances
kern_fun	kernel function to use
p1	index of X to permute
reg_threshold	norm constraint for model class
tol	tolerance used for calculations
nrep_sample	see get_full_sample
verbose	whether to report results during computation
warn_psd	whether to warn that small numbers are added to make the kernel matrix positive definite
warn_duplicate	whether to warn if duplicated rows of dat_ref are dropped
warn_dropped	passed to get_full_sample

aaronjfisher/mcr documentation built on Jan. 2, 2020, 4:38 p.m.