tsnet (version 0.2.0)

post_distance_within: Calculates distances between pairs of posterior samples using the posterior samples or posterior predictive draws

Description

This function computes distances between posterior samples of a single fitted GVAR model. Thereby, it calculates the uncertainty contained in the posterior distribution, which can be used as a reference to compare two modes. Distances can be obtained either from posterior samples or posterior predictive draws. The distance between two models can currently be calculated based on three options: Frobenius norm, maximum difference, or L1 norm. Used within compare_gvar. The function is not intended to be used directly by the user.

Usage

post_distance_within(
  fitobj,
  comp,
  pred,
  n_draws = 1000,
  sampling_method = "random",
  indices = NULL,
  burnin = 0
)

Value

A list of distances between the specified pairs of fitted models. The list has length equal to the specified number of random pairs. Each list element contains two distance values, one for beta coefficients and one for partial correlations.

Arguments

fitobj

Fitted model object. This can be a tsnet_fit object (obtained from stan_gvar, a BGGM object (obtained from var_estimate, or extracted posterior samples (obtained from stan_fit_convert).

comp

The distance metric to use. Should be one of "frob" (Frobenius norm), "maxdiff" (maximum difference), or "l1" (L1 norm) (default: "frob"). The use of the Frobenius norm is recommended.

pred

A logical indicating whether the input is posterior predictive draws (TRUE) or posterior samples (FALSE). Default: FALSE

n_draws

The number of draws to use for reference distributions (default: 1000).

sampling_method

Draw sequential pairs of samples from the posterior, with certain distance between them ("sequential") or randomly from two halves of the posterior ("random"). The "random" method is preferred to account for potential autocorrelation between subsequent samples. Default: "random".

indices

A list of "beta" and "pcor" indices specifying which elements of the matrices to consider when calculating distances. If NULL (default), all elements of both matrices are considered. If provided, only the elements at these indices are considered. If only one of the matrices should have indices, the other one should be NULL. This can be useful if you want to calculate distances based on a subset of the elements in the matrices.

burnin

The number of burn-in iterations to discard (default: 0).

Examples

data(fit_data)
post_distance_within(fitobj = fit_data[[1]],
comp = "frob",
pred = FALSE,
n_draws = 100)