Package: cevcmm 0.1.4

Lida Jalili

cevcmm: Communication-Efficient Varying Coefficient Mixed-Effects Models

Scalable inference for Varying Coefficient Mixed-Effects Models (VCMMs) with large, correlated random effects. Implements sufficient-statistics, one-step communication-efficient surrogate likelihood, and SVD-stabilized (Singular Value Decomposition) estimators with Kronecker and separable covariance structures. Implements the methodology of Jalili and Lin (2025) <doi:10.48550/arXiv.2511.12732>.

Authors:Lida Jalili [aut, cre], Li-Hsiang Lin [aut]

cevcmm_0.1.4.tar.gz
cevcmm_0.1.4.zip(r-4.7-x86_64)cevcmm_0.1.4.zip(r-4.7-arm64)cevcmm_0.1.4.zip(r-4.6-x86_64)cevcmm_0.1.4.zip(r-4.6-arm64)cevcmm_0.1.4.zip(r-4.5-x86_64)
cevcmm_0.1.4.tgz(r-4.6-x86_64)cevcmm_0.1.4.tgz(r-4.6-arm64)cevcmm_0.1.4.tgz(r-4.5-x86_64)cevcmm_0.1.4.tgz(r-4.5-arm64)
cevcmm_0.1.4.tar.gz(r-4.7-arm64)cevcmm_0.1.4.tar.gz(r-4.7-x86_64)cevcmm_0.1.4.tar.gz(r-4.6-arm64)cevcmm_0.1.4.tar.gz(r-4.6-x86_64)
cevcmm_0.1.4.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
cevcmm/json (API)

# Install 'cevcmm' in R:
install.packages('cevcmm', repos = c('https://lidajalili.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/lidajalili/cevcmm/issues

Pkgdown/docs site:https://lidajalili.github.io

Uses libs:
  • openblas– Optimized BLAS
  • c++– GNU Standard C++ Library v3

On CRAN:

Conda:

openblascpp

5.15 score 1 stars 19 scripts 221 downloads 18 exports 2 dependencies

Last updated from:13ed100df8. Checks:15 OK. Indexed: yes.

TargetResultTimeFilesSyslog
linux-devel-arm64OK138
linux-devel-x86_64OK165
source / vignettesOK271
linux-release-arm64OK132
linux-release-x86_64OK158
macos-release-arm64OK203
macos-release-x86_64OK338
macos-oldrel-arm64OK180
macos-oldrel-x86_64OK345
windows-devel-arm64OK378
windows-devel-x86_64OK180
windows-release-arm64OK363
windows-release-x86_64OK111
windows-oldrel-x86_64OK131
wasm-releaseOK143

Exports:accumulate_statsbuild_kronecker_precisionbuild_penalty_matrixbuild_vcmm_designcevcmm_rcpp_checkcompute_sufficient_statsestimate_kronecker_componentsfit_cslfit_from_summariesfit_ssfixefinit_accumulatorinvert_matrixnode_summaryranefvarying_coefvcmmvcmm_control

Dependencies:RcppRcppArmadillo

Distributed fitting with cevcmm
When you need this | The math guarantee | Setup | Simulate | Pattern 1 — list of node summaries | Verify bit-equivalence against pooled fitting | Pattern 2 — streaming accumulator | Identifiability for OD designs | Where to go next | Reference
Rendered fromdistributed-fitting.Rmdusingknitr::rmarkdown

Last update: 2026-07-06
Started: 2026-06-30

Getting started with cevcmm
What this package fits | Setup | Simulate a small dataset | Fit | Inspect the fit | Recover the varying coefficient | Predict on new data | Built-in diagnostic plots | Where to go next | Reference
Rendered fromgetting-started.Rmdusingknitr::rmarkdown

Last update: 2026-07-06
Started: 2026-06-30

OD migration with Kronecker covariance
The OD setting | Setup | Load the bundled example | Build the random-effects design | Choose an initial spatial covariance | Fit | Interpret the estimated $\Sigma_{2 \times 2}$ | Recover the per-region random effects | Estimated varying coefficient | Distributing this fit | Where to go next | Reference
Rendered fromod-migration.Rmdusingknitr::rmarkdown

Last update: 2026-07-06
Started: 2026-06-30

Readme and manuals

Help Manual

Help pageTopics
Additive aggregation of sufficient-statistics summaries+.vcmm_ss
Add one batch or node statistics to a running accumulatoraccumulate_stats
Build the Kronecker-structured random-effect precision matrixbuild_kronecker_precision
Build a B-spline second-order difference penalty matrixbuild_penalty_matrix
Build the design matrix and penalty for a VCMMbuild_vcmm_design
Fixed-effects coefficient vector from a vcmm fitcoef.vcmm_fit
Compute one batch or node sufficient statistics for a normal linear VCMMcompute_sufficient_stats
Moment-based estimator for the Kronecker left componentestimate_kronecker_components
Fit a VCMM via the one-step CSL estimatorfit_csl
Fit a VCMM from aggregated sufficient-statistics summariesfit_from_summaries
Fit a VCMM via the sufficient-statistics estimatorfit_ss print.vcmm_fit
Extract fixed-effects from a fitted model objectfixef
Fixed effects of a VCMM, reshaped by varying-coefficient blockfixef.vcmm_fit
Initialize an empty sufficient-statistics accumulatorinit_accumulator
Numerically stable matrix inversion with automatic method selectioninvert_matrix
Log-likelihood of a vcmm fitlogLik.vcmm_fit
Number of observations from a vcmm fitnobs.vcmm_fit
Compute one node's sufficient-statistics summarynode_summary
Diagnostic plots for a vcmm fitplot.vcmm_fit
Predictions from a fitted VCMMpredict.vcmm_fit
Extract random effects from a fitted model objectranef
Random effects of a VCMM, reshaped by re_cov structureranef.vcmm_fit
Summarise a vcmm fitprint.vcmm_summary summary.vcmm_fit
Evaluate the varying coefficient(s) at new t valuesvarying_coef
Fit a Varying Coefficient Mixed-Effects Modelvcmm
Control parameters for VCMM fittingprint.vcmm_control vcmm_control
Variance-covariance matrix of the fixed-effects from a vcmm fitvcov.vcmm_fit