Collect stLMM posterior samples in a data frame

Collect posterior sample matrices from fitted, recovered, prediction, or fitted-value objects into a plain data frame with samples in rows and named sample columns across the top.

Usage

as_samples(object, ...)

Arguments

object: An stLMM, stLMM_chains, stLMM_recovery, stLMM_recovery_chains, stLMM_prediction, stLMM_prediction_chains, stLMM_fitted_chains, or posterior sample matrix.
...: Additional arguments passed to methods. Supported arguments include burn, thin, metadata, include_w, combine_chains, and, for prediction objects, sample.

Details

For fitted stLMM objects, columns contain active parameter sample blocks such as fixed effects, grouped random effects, residual variance parameters, process variances, and process correlation parameters. Structured latent process draws are available when they were saved during fitting or added by recover(), and are included only when include_w = TRUE.

For recovery objects, parameter samples are aligned to object$recover_iter before saved or recovered latent process draws are appended. This ensures parameter columns and w_ columns describe the same retained posterior iterations.

For multi-chain objects, samples are combined by row by default and the .chain metadata column preserves chain membership. Set combine_chains = FALSE to return one data frame per chain.

Prediction methods accept sample = "mu", "y", or "all". Prediction columns are named mu_1, mu_2, and so on, or y_1, y_2, and so on.

Value

A data frame, or a list of data frames when combine_chains = FALSE. When metadata = TRUE, the first columns are .chain and .iteration. .iteration is the original MCMC iteration within chain when that metadata is available.

Examples

set.seed(1)
dat <- data.frame(y = rnorm(8), x = seq(-1, 1, length.out = 8))
fit <- stLMM(y ~ x, data = dat, n_samples = 8, warmup = FALSE, verbose = FALSE)
draws <- as_samples(fit, burn = 2, thin = 2)

newdat <- data.frame(x = c(-0.5, 0.5))
pred <- predict(fit, newdata = newdat, y_samples = TRUE)
pred_draws <- as_samples(pred, sample = "all")