Predict at new sites from a fitted Bayesian hierarchical spatial model

Bayesian kriging: for each posterior draw of (beta, sigma^2, tau^2) we sample the latent GP at new locations conditional on the observed latent field, then add iid noise. Returns posterior mean and quantiles at each newdata row.

Usage

# S3 method for class 'edaphos_bhs'
predict(object, newdata, quantiles = c(0.025, 0.5, 0.975), n_draws = 500L, ...)

Arguments

object

An edaphos_bhs fit.

newdata

A data frame with covariates + coordinates matching the training schema.

quantiles

Quantile levels to return. Default c(0.025, 0.5, 0.975).

n_draws

Integer; how many posterior samples to use (capped by the actual number available in the fit). Default 500L.

...

Unused.

Value

A data frame with newdata rows plus columns mean, sd, and one column per quantile.