Kernel Stein Discrepancy Goodness-of-Fit and Stein Sampling Tools

---

Documentation for package ‘steinsampling’ version 0.1.0

• DESCRIPTION file

Help Pages

steinsampling-package	steinsampling: Stein tests and Stein sampling tools
compute_fssd_null_pvalue	Simulate the FSSD null distribution
compute_fssd_unbiased_stat	Estimate squared FSSD from a feature matrix
compute_tau	Compute the FSSD feature matrix
cross_kernel	Building blocks for Stein kernels
custom_adjusted_gradient	Apply a custom SVGD direction adjustment
custom_stein_kernel	Create a custom Stein kernel
eval_kernel	Building blocks for Stein kernels
find_median_distance	Median squared distance between sample pairs
fmin_grid	Create the grid search used by Stein Points
fmin_mc	Create the Monte Carlo search used by Stein Points
fmin_nm	Create the multi-start Nelder-Mead search used by Stein Points
fssd_opt_test	FSSD test with optimized test locations
fssd_rand_test	FSSD test with random test locations
fssd_test	Finite Set Stein Discrepancy goodness-of-fit test
get_score_evaluator	Create a score function for a fixed Gaussian mixture model
gmm	Create a Gaussian mixture model
grad_theta_v_kernel	Building blocks for Stein kernels
grad_x_kernel	Building blocks for Stein kernels
grw	Run the Gaussian random-walk transition used by SP-MCMC
grwmetrop	Run a Gaussian random-walk Metropolis chain
kernel_generics	Building blocks for Stein kernels
ksd_uq_matrix	Build the Stein-kernel matrix for the KSD U-statistic
ksd_u_bootstrap	Bootstrap the KSD U-statistic from a Stein-kernel matrix
ksd_u_statistic	Compute the KSD U-statistic from its Stein-kernel matrix
ksd_u_test	KSD goodness-of-fit test using the Liu et al. U-statistic
ksd_vq_matrix	Build the Stein-kernel matrix for the KSD V-statistic
ksd_v_bootstrap	Wild-bootstrap the KSD V-statistic
ksd_v_statistic	Compute the KSD V-statistic from its Stein-kernel matrix
ksd_v_test	KSD goodness-of-fit test using the Chwialkowski et al. V-statistic
likelihoodgmm	Compute the mixture density for samples
mala	Run a Metropolis-adjusted Langevin chain
perturbgmm	Perturb the component means of a Gaussian mixture model
plotgmm	Plot a one-dimensional Gaussian mixture sample
posteriorgmm	Compute posterior component probabilities for a Gaussian mixture
rgmm	Sample from a Gaussian mixture model
rwm	Alias for the Gaussian random-walk transition
scorefunctiongmm	Compute the score of a Gaussian mixture density
sp_mcmc	Generate Stein Points from short MCMC candidate paths
sp_mcmc_criterion	Create an SP-MCMC start-point rule
sp_mcmc_eval_candidates	Score SP-MCMC candidate points
sp_mcmc_select_start	Choose the next SP-MCMC chain start
sp_mcmc_state	Store the current SP-MCMC state for a start rule
steinsampling	steinsampling: Stein tests and Stein sampling tools
stein_codescent	Refine Stein Points by coordinate descent
stein_kernel	Create a built-in Stein kernel
stein_kernel_imq_score	Create a score-distance IMQ Stein kernel
stein_kernel_inverse_log	Create an inverse-log Stein kernel
stein_kernel_matrix	Compute pairwise Stein kernel values
stein_points	Build a Stein Points sequence by greedy KSD minimization
stein_thinning	Compress existing samples with Stein thinning
svgd	Create an SVGD updater
trace_mixed_kernel	Building blocks for Stein kernels
update_svgd	Run SVGD particle updates