Skip to contents

Summarize Scalar Parameters

Source: R/posterior_extract.R
param_summary.Rd

Returns posterior mean, standard deviation, and quantiles for the scalar variance parameters estimated by the model. These typically include:

  • sigma2 or s2: process noise variance

  • tau_A2 / tau_B2 or t2: innovation variance for A/B

  • g2: latent variance

  • rhoA / rhoB: AR(1) persistence (dynamic model with ar1 = TRUE)

  • sigma2_obs: observation variance (gaussian family)

Usage

param_summary(fit, probs = c(0.025, 0.5, 0.975))

Arguments

fit

A dbn model fit object (output from dbn())

probs

Quantile probabilities (default: 5th, 50th, 95th percentiles)

Value

Data frame with columns: parameter, mean, sd, and one column per requested quantile

See also

theta_summary, plot_trace, derive_draws

Examples

# \donttest{
sim <- simulate_dynamic_dbn(n = 6, time = 5, seed = 1)
fit <- dbn(sim$Y, model = "dynamic", nscan = 200, burn = 100, verbose = FALSE)
ps <- param_summary(fit)
# }