ed.stan.stanfit: Empirical determinacy estimates for Stan

In hunansona/ed4bhm: Empirical Determinacy for Bayesian Hierarchical Models

Empirical determinacy estimates for Stan

Description

Provides empirical determinacy estimates TED, EDL, EDS, pEDL, and pEDS from a Stan model. This function takes the Stan fit object from the base model as input and returns the identification values, descritive.matrix and delta.

Usage

ed.stan(stan.object.base,
                    prec.name = NULL, 
                    delta = 0.01,
                    distance = "H2ALL")

Arguments

stan.object.base	Stan fit object from the base model with a column called "ll", which contains the log-likelihood evaluated at MCMC sample of parameters.
prec.name	NULL or character, specifies the names of parameters (in JAGS and INLA precisions but in Stan standard deviations) to be log-transformed
delta	numeric, the numerical differentiation step and w = 1 \pm δ is the weight, default value is 0.01.
distance	character string, specifies the type of the empirical determinacy measure. It can have values "H2" for TED, "BCL" for EDL and "BCS" for EDS and "H2ALL" for all the five measures (TED, EDL, EDS, pEDL and pEDS). The default value is "H2ALL".

Value

H2ALL measures	a matrix of identification values
descriptive matrix	A list of descriptive matrix and delta

Author(s)

Malgorzata Roos, Georgios Kazantzidis, Sona Hunanyan,

hunansona/ed4bhm documentation built on June 15, 2022, 6:42 p.m.