README

cusna (native) — simulation and estimation of SAOM/RSiena, ERGM, and friends for R

cusna is a self-contained native engine — a small C interface over C++ host logic (libcusna) — callable from R without a Python runtime:

the R package builds the engine’s CPU backend from source (C++17, multi-threaded with OpenMP; cusna_set_threads() sets the thread count);

a CUDA backend implements the same C interface and is built separately from the native/ sources (it is driven by the companion Python package); cusna_has_cuda() reports FALSE for the R package.

What the package provides

Family	Functions	Validated against
SAOM (RSiena)	`saom_data()`, `cusna_effect()`, `mom_estimate()` (with `return_nets`), `cusna_fit` methods, `saom_simulate()`; behavior co-evolution, composition change, `mom_estimate_multinet()`, `cusna_fran()` for `siena07()`	data/masks/targets bit-identical to the reference; estimates within simulation SE; RSiena targets to machine zero
ERGM	`ergm_simulate()` (TNT sampler), `ergm_stats()`, `ergm_mple()`, `ergm_mcmle()`	sampler ≡ `ergm::simulate`; MLE matches `ergm::ergm()`
Temporal ERGM	`tergm_mple()` (+ block bootstrap), `tergm_simulate()`, `stergm_cmle()`, `stergm_simulate()`	matches `btergm` to machine precision; `tergm` CMLE within SE
ALAAM	`alaam_mple()`, `alaam_mcmle()`, `alaam_simulate()`	MPLE ≡ `glm`; MLE recovers observed moments
Low-level	`cusna_network_stats()`, `cusna_behavior_stats()`, `cusna_gof_distribution()`	RSiena Appendix B conventions, machine zero

Family

Functions

Validated against

SAOM (RSiena)

saom_data(), cusna_effect(), mom_estimate() (with return_nets), cusna_fit methods, saom_simulate(); behavior co-evolution, composition change, mom_estimate_multinet(), cusna_fran() for siena07()

data/masks/targets bit-identical to the reference; estimates within simulation SE; RSiena targets to machine zero

ERGM

ergm_simulate() (TNT sampler), ergm_stats(), ergm_mple(), ergm_mcmle()

sampler ≡ ergm::simulate; MLE matches ergm::ergm()

Temporal ERGM

tergm_mple() (+ block bootstrap), tergm_simulate(), stergm_cmle(), stergm_simulate()

matches btergm to machine precision; tergm CMLE within SE

ALAAM

alaam_mple(), alaam_mcmle(), alaam_simulate()

MPLE ≡ glm; MLE recovers observed moments

Low-level

cusna_network_stats(), cusna_behavior_stats(), cusna_gof_distribution()

RSiena Appendix B conventions, machine zero

The underlying C interface is validated bit-for-bit in native/test (see native/VALIDATION.md).

Quickstart

install.packages("cusna")
library(cusna)

# the bundled s50 panel: three waves of friendship among 50 girls
dat <- saom_data(s50$waves)
fit <- mom_estimate(dat,
                    effects = list(cusna_effect("density"), cusna_effect("recip"),
                                   cusna_effect("transTrip")),
                    conditional = TRUE, return_nets = TRUE)
summary(fit)

# goodness of fit on the phase-3 networks: simulated indegree distribution
cusna_gof_distribution(fit$sims[[2]], kind = 0L, levls = 0:6)

# an ERGM maximum-likelihood fit on the first wave
ergm_mcmle(s50$waves[[1]], list(ergm_term("edges"), ergm_term("mutual")),
           directed = TRUE)

See vignette("cusna") for the full tour (covariates, co-evolution, multi-network models, TERGM/STERGM/ALAAM) and vignette("siena07-backend") for driving RSiena’s siena07() on the native simulator.

Building from the repository

See BUILD.md. In short: the package needs a C++17 compiler (Rtools on Windows); its configure script vendors the native sources from ../native/.

License

MIT (see LICENSE). We compare outputs against RSiena but do not link its code.