Skip to contents

Attach posterior draws to an Opal object

opal_attach_mcmc.Rd

Validate and store posterior draws for the object's current configuration. Parameter names must identify the active model layout. Matrices contain draws in rows and named variables in columns; arrays use iteration, chain, and variable dimensions. Use opal_mcmc() to run a new sampler.

Usage

opal_attach_mcmc(x, mcmc, settings = list(), check = TRUE, check_args = list())

Arguments

x

A configured or fitted opal_obj.

mcmc

SparseNUTS output, sample matrix, or iteration-chain-variable array.

settings

Named portable sampler settings.

check

Run posterior diagnostics.

check_args

Arguments passed to opal_check().

Value

An updated opal_obj. Previous selected draws are retained in history.

Details

A failed or unchecked import does not replace a previously checked, passing posterior; inspect x$mcmc_history for unselected attempts. Parameter-only imports lack sampler diagnostics, so a passing MCMC check requires richer sampler output. Full joint draws are required for random-effect projections.

See also

Other assessment workflow: opal_attach_fit(), opal_build(), opal_check(), opal_derived(), opal_example_inputs(), opal_fit(), opal_from_fit(), opal_io, opal_mcmc(), opal_obj(), opal_posterior(), opal_project(), opal_report(), opal_update()