Kullback-Leibler Optimal Gaussian Mixture Proxies for Target Densities

Documentation for package ‘proxymix’ version 0.16.0

Help Pages

-- A --

as_mids	Convert imputations to a mice multiply-imputed dataset
autoplot.gmm_fit	Plot a fitted Gaussian-mixture proxy

-- B --

banana_target	Banana-shaped 2-D target
bic_aic	Information criteria: BIC, AIC, and ICL

-- C --

censored

Missingness mechanisms for multiple imputation

-- D --

dgmm	Density of a Gaussian mixture
donut_target	Donut-shaped 2-D target

-- E --

epanechnikov_target	Compact-support Epanechnikov target
ess_summary	Summary of importance-sampling diagnostics
ess_trace	Effective sample size of the importance-sampling weights

-- F --

fit_em_samples	Classical EM fit on samples
fit_kld_em	Importance-sampled KLD-EM fit (regime iii)
fit_moment_match	Closed-form moment-matching fit
fit_proxymix	Fit a Gaussian-mixture proxy to a target density
fit_uplift	Fit an uplift / next-best-action model from a data frame
from_kde	Compile a kernel-density estimate into a Gaussian-mixture proxy
from_objective	Map the optima of an objective with a Gaussian-mixture proxy

-- G --

glance.gmm_fit	Glance at a fitted Gaussian-mixture proxy
gmm	A Gaussian mixture
gmm_affine	Affine pushforward of a Gaussian mixture
gmm_aggregate	Aggregation pushforward of a Gaussian mixture
gmm_anneal_path	Phase-transition component discovery by deterministic annealing
gmm_canonicalise	Canonicalise the component ordering of a Gaussian mixture
gmm_cf_mean	The identified counterfactual mean
gmm_cf_tail_prob	Refused: a tail probability of an individual counterfactual law
gmm_cf_variance	Refused: the variance of an individual counterfactual law
gmm_complete	Extract completed datasets from a 'gmm_imputation'
gmm_conditionalise	Conditional of a Gaussian mixture
gmm_conditional_entropy	Renyi-2 or Shannon entropy of a conditional Gaussian mixture
gmm_convolve	Convolution of two independent Gaussian mixtures
gmm_counterfactual	Counterfactual law of one unit (abduction, action, prediction)
gmm_counterfactual_law	A per-unit counterfactual law
gmm_cov	Mean and covariance of a Gaussian mixture
gmm_covariances	Component parameters of a Gaussian mixture
gmm_dim	Dimension of a Gaussian mixture
gmm_divergence	Divergence between two Gaussian mixtures
gmm_entropy	Renyi-2 or Shannon entropy of a Gaussian mixture
gmm_eos_test	End-of-sample instability test on a Gaussian state-space filter
gmm_evidence	Estimate the target's normalising constant from a fitted proxy
gmm_filter	Bounded Gaussian-sum filtering over an observation series
gmm_fit	A fitted Gaussian-mixture proxy
gmm_fit_ensemble	Bootstrap ensemble of a fitted proxy
gmm_fit_quality	The quality certificate of a fit or derived mixture
gmm_imputation	A Gaussian-mixture multiple-imputation result
gmm_impute	Multiple imputation by Gaussian-mixture conditioning
gmm_independence_graph	Conditional-independence (Gaussian graphical model) structure of a mixture
gmm_intervene	Interventional law of a Gaussian mixture (the do-operator)
gmm_kld	Kullback-Leibler divergence between two Gaussian mixtures
gmm_marginalise	Marginal of a Gaussian mixture
gmm_mean	Mean and covariance of a Gaussian mixture
gmm_means	Component parameters of a Gaussian mixture
gmm_missing	Condition a Gaussian mixture on the exact values of some coordinates
gmm_mix	Mix Gaussian mixtures into one mixture
gmm_modes	Modes of a Gaussian mixture
gmm_mutual_information	Cauchy-Schwarz mutual information between two coordinate blocks
gmm_n_components	Number of components in a Gaussian mixture
gmm_observe	Bayesian update of a Gaussian mixture on a noisy linear observation
gmm_product	Pointwise product of two Gaussian mixtures
gmm_reduce	Reduce a Gaussian mixture to fewer components
gmm_target	A target density on R^p
gmm_target_from_posterior	Compile an unnormalised Bayesian posterior into a 'gmm_target'
gmm_target_from_posterior.default	Compile an unnormalised Bayesian posterior into a 'gmm_target'
gmm_target_from_posterior.function	Compile an unnormalised Bayesian posterior into a 'gmm_target'
gmm_target_from_samples	Build a target from samples alone
gmm_weights	Component parameters of a Gaussian mixture

-- H --

hellinger_mc

Monte-Carlo Hellinger distance between a fit and its target

-- I --

init_kmeans	k-means initialisation
init_moment_seed	Moment-seed initialisation
init_random	Random initialisation
init_warm_start	Warm-start initialisation from an existing fit
is_mvn	Multivariate-normal proposal
is_mvt	Multivariate-t proposal
is_proposal	An importance-sampling proposal
is_uniform	Uniform-on-a-box proposal

-- K --

kld_trace

Per-iteration KLD trace of a fit

-- M --

mar	Missingness mechanisms for multiple imputation
maxent_target	Maximum-entropy target under moment and support constraints
mechanism	Missingness mechanisms for multiple imputation
mixture_target	Three-component Gaussian-mixture target
mnar	Missingness mechanisms for multiple imputation
multi_start_best_of	Multi-start best-of wrapper

-- P --

pgmm	Distribution and quantile functions of a one-dimensional mixture
proposal_mvn	Preferred names for the importance-proposal constructors
proposal_mvt	Preferred names for the importance-proposal constructors
proposal_uniform	Preferred names for the importance-proposal constructors
proxy_cate	Heterogeneous treatment effects (CATE / uplift)
proxy_confounding_gap	Confounding gap: the sensitivity of the effect to the latent regime
proxy_decide	Optimal action and expected incremental value per unit
proxy_fmi	Fraction of missing information for a column mean
proxy_functional_ci	Percentile interval for any functional of a fitted proxy
proxy_identification_report	The identification report (an executive one-pager)
proxy_mnar_sensitivity	Missing-not-at-random sensitivity analysis for a coordinate mean
proxy_overlap	Per-unit overlap / positivity diagnostic
proxy_policy_value	Off-line value of a targeting policy
proxy_pool	Pool a column mean across imputations
proxy_predict	Predicted outcome under a treatment (the seeing rung)
proxy_regime_segments	The fitted regimes as an interpretable segment table
proxy_retrospective_uplift	Retrospective (counterfactual-mean) uplift for observed units
proxy_uplift	Uplift (alias of 'proxy_cate()' for a binary treatment)

-- Q --

qgmm	Distribution and quantile functions of a one-dimensional mixture

-- R --

rgmm	Sample from a Gaussian mixture

-- S --

select_N

Select the number of mixture components

-- T --

tidy.gmm

Tidy a Gaussian mixture into a component table

-- U --

uplift_identification	Identification-report object
uplift_model	A fitted uplift / next-best-action model