Package index • cevcmm

Main fit function

Single-call entry point for all three estimators and covariance modes.

vcmm(): Fit a Varying Coefficient Mixed-Effects Model
vcmm_control() print(<vcmm_control>): Control parameters for VCMM fitting

Per-node summaries that aggregate additively, plus the central fit function that consumes them.

node_summary(): Compute one node's sufficient-statistics summary
compute_sufficient_stats(): Compute one batch or node sufficient statistics for a normal linear VCMM
init_accumulator(): Initialize an empty sufficient-statistics accumulator
accumulate_stats(): Add one batch or node statistics to a running accumulator
`+`(<vcmm_ss>): Additive aggregation of sufficient-statistics summaries
fit_from_summaries(): Fit a VCMM from aggregated sufficient-statistics summaries

Direct access to the SS and CSL fit functions, called internally by vcmm(). Advanced users only.

fit_ss() print(<vcmm_fit>): Fit a VCMM via the sufficient-statistics estimator
fit_csl(): Fit a VCMM via the one-step CSL estimator

B-spline design matrix and difference-penalty construction.

build_vcmm_design(): Build the design matrix and penalty for a VCMM
build_penalty_matrix(): Build a B-spline second-order difference penalty matrix

Helpers for the Kronecker / separable covariance machinery.

build_kronecker_precision(): Build the Kronecker-structured random-effect precision matrix
estimate_kronecker_components(): Moment-based estimator for the Kronecker left component

Stabilised matrix inversion used inside the estimators.

invert_matrix(): Numerically stable matrix inversion with automatic method selection
svd_pseudo_inverse(): SVD-based Moore-Penrose pseudo-inverse
split_merge_svd_row(): Split-merge SVD via row partitioning
cevcmm_rcpp_check(): Verify the Rcpp / RcppArmadillo toolchain is wired up

Standard generics for inspecting and using a vcmm_fit.

coef(<vcmm_fit>): Fixed-effects coefficient vector from a vcmm fit
fixef(): Extract fixed-effects from a fitted model object
fixef(<vcmm_fit>): Fixed effects of a VCMM, reshaped by varying-coefficient block
ranef(): Extract random effects from a fitted model object
ranef(<vcmm_fit>): Random effects of a VCMM, reshaped by re_cov structure
vcov(<vcmm_fit>): Variance-covariance matrix of the fixed-effects from a vcmm fit
nobs(<vcmm_fit>): Number of observations from a vcmm fit
logLik(<vcmm_fit>): Log-likelihood of a vcmm fit
summary(<vcmm_fit>) print(<vcmm_summary>): Summarise a vcmm fit
predict(<vcmm_fit>): Predictions from a fitted VCMM
plot(<vcmm_fit>): Diagnostic plots for a vcmm fit
varying_coef(): Evaluate the varying coefficient(s) at new t values

cevcmm cevcmm-package: cevcmm: Communication-Efficient Varying Coefficient Mixed-Effects Models