Function reference • varapproxr

All functions

ExpectationGaussianSufficientStatistics(): Gaussian transform natural parameters to E[T(X)] parameters.

ExpectationInverseGWishartSufficientStatistics(): Inverse-G-Wishart transform natural parameters to E[T(X)] parameters.

G_VMP(): Calculate

GaussianCommonParameters(): Gaussian transform natural parameters to E[T(X)] parameters.

GaussianEntropy(): Calculate entropy for Gaussian density

GaussianLikelihoodFragment(): Gaussian likelihood fragment update

GaussianPriorFragment(): Gaussian prior fragment update

InverseGWishartCommonParameters(): Inverse-G-Wishart Common parameters

InverseGWishartPriorFragment(): Inverse G-Wishart prior fragment update

InverseGammaPriorFragment(): Inverse-Gamma prior fragment update

InverseWishartPriorFragment(): Inverse Wishart prior fragment update

IteratedInverseGWishartFragment(): Iterated Inverse G-Wishart fragment update

arma2vec(): Convert arma::vec to Rcpp::NumericVector

bind_cols(): Right-bind columns of matrices from in list.

blockDiag(): Construct block-diagonal matrix from list of matrices

check.mfvb.x(): Check the `x` input to an mfvb regression function

check.mfvb.y(): Check the `y` input to an mfvb regression function

coef(<mfvb>): Extract model coefficients posterior means

confint(<mfvb>): Compute credible intervals for parameters of `mfvb`

dnorm_mat(): Evaluate standard normal density for matrix of variates

dot_y_minus_Xb(): E[||y - Xb||^2]

ig_E(): Inverse Gamma E[x]

ig_E_inv(): Inverse Gamma E[1/x]

ig_E_log(): Inverse Gamma E[log(x)]

ig_E_lpdf(): Calculate E_q[ln p(x)] where q(x) = IG(x | a, b) and p(x) = IG(x | a0, b0)

ig_entropy(): Inverse Gamma H[x]

inv_vech(): Calculate inverse of vech for a vector

inv_vectorise(): Calculate vec^(-1)

inv_wishart_E_invX(): Inverse Wishart E[X^-1]

inv_wishart_E_logdet(): Inverse Wishart E[log|X|]

jaakkola_jordan(): Perform Jaakkola-Jordan update of variational parameters

jaakkola_jordan_n(): Perform Jaakkola-Jordan update of variational parameters

knowles_minka_wand(): Perform Knowles-Minka-Wand update of variational parameters

knowles_minka_wand_n(): Perform Knowles-Minka-Wand update of variational parameters

lmvgamma(): Log multivariate Gamma function

mfvb.control(): Control arguments for mfvb functions

mfvb_lm(): #' mfvb_lm.fit #' #' @param x The x input #' @param y The response #' @param weights weighting #' @param subset subset of data #' @param na.action handling of na #' @param offset Offset term #' @param control Control terms #' #' @return x if all checks pass mfvb_lm.fit <- function( x = stop("no 'x' argument"), y = stop("no 'y' argument"), weights = NULL, subset = NULL, na.action = na.fail, offset = NULL, control = list(), ... ) control <- do.call("mfvb.control", control) x <- check.mfvb.x(x) y <- check.mfvb.y(x, y) mfvb_lm

mvdigamma(): Multivariate digamma function

mvgamma(): multivariate Gamma function

mvn_E_lpdf(): Calculate E_q[ln p(x)] where q(x) = MVN(x | mu, Sigma) and p(x) = MVN(x | mu0, Sigma0)

mvn_entropy(): Multivariate Normal Entropy H[x]

ph_exponential(): Normal parametric variational Bayes for Exponential PH Model.

pnorm_mat(): Evaluate standard normal cdf for matrix of variates

sample_vbdist(): Sample from MFVB fit

saul_jordan(): Perform Saul-Jordan update of variational parameters

saul_jordan_n(): Perform Saul-Jordan update of variational parameters

scaled_inv_chisq_E(): Scaled Inverse Chi squared E[x]

scaled_inv_chisq_E_inv(): Scaled Inverse Chi squared E[1/x]

scaled_inv_chisq_E_log(): Scaled Inverse Chi squared E[log(x)]

scaled_inv_chisq_H(): Scaled Inverse Chi squared H[x]

solve_two_level_sparse(): Solve two level sparse matrix problem.

summary(<mfvb>): Summarise MFVB fit

update_vb_lm(): Update mean-field variational inference for a normal linear model.

varapproxr-package varapproxr: Variational approximations

vb_glmm(): Variational Bayes for logistic mixed model.

vb_lm(): Mean-field variational inference for a normal linear model.

vb_lmm(): Variational Bayes for linear mixed model.

vb_lmm_randint(): Variational Bayes for linear mixed model (random intercept only).

vb_lmm_randintslope(): Variational Bayes for linear mixed model (random intercept and coefficient only).

vb_lmm_randintslope_streamlined(): Variational Bayes for linear mixed model (random intercept and coefficient only, streamlined).

vb_logistic(): Perform variational inference for logistic regression model

vb_logistic_n(): Variational inference for binomial logistic regression model

vb_pois_mm(): Poisson Mixed Effects Model

vb_pois_reg(): Poisson regression

vcov(<mfvb>): Extract model coefficient posterior covariance matrix

vech(): Calculate vech of a matrix

vmp_lm(): Variational Message Passing for Normal linear model.

woodbury(): Woodbury matrix identity