simulate.tulpa_fit: Simulate responses from a fitted tulpa model

In tulpa: Templated Unified Library for Posterior Approximation in Bayesian Hierarchical Models

Simulate responses from a fitted tulpa model

Description

Base-R alias for posterior_predict(): each simulation is one posterior predictive replicate at the training data. A categorical fit (tulpa_multinomial(), tulpa_ordinal()) simulates factor columns carrying the response levels.

Usage

## S3 method for class 'tulpa_fit'
simulate(object, nsim = 1, seed = NULL, ...)

Arguments

object	A tulpa_fit object.
nsim	Number of simulated datasets (default 1).
seed	Optional integer seed (RNG state is restored on exit).
...	Ignored.

Value

A data frame with nsim columns (sim_1, ...), one row per observation, following the stats::simulate() convention: its "seed" attribute is the value of seed (with the RNG kind as attribute "kind") when one was given, else the state of .Random.seed before simulation.

tulpa documentation built on Oct. 19, 2026, 5:09 p.m.