blavaan (version 0.1-4)

dpriors: Specify default prior distributions

Description

Specify "default" prior distributions for classes of model parameters.

Usage

dpriors(nu = "dnorm(0,1e-3)", alpha = "dnorm(0,1e-2)",  lambda = "dnorm(0,1e-2)", beta = "dnorm(0,1e-2)",  itheta = "dgamma(1,.5)", ipsi = "dgamma(1,.5)",  rho = "dbeta(1,1)", ibpsi = "dwish(iden,3)")

Arguments

nu

Prior distribution for nu (observed variable intercept) parameters.

alpha

Prior distribution for alpha (latent variable intercept) parameters.

lambda

Prior distribution for lambda (loading) parameters.

beta

Prior distribution for regression parameters.

itheta

Prior distribution for observed variable precision parameters.

ipsi

Prior distribution for latent variable precision parameters.

rho

Prior distribution for correlation parameters (only used under srs approach).

ibpsi

Prior distribution for inverse covariance matrix of blocks of latent variables.

Value

A character vector containing the prior distribution for each type of parameter.

Details

User-specified prior distributions for specific parameters (using the prior() operator within the model syntax) always override prior distributions set using dpriors().

References

Edgar C. Merkle & Yves Rosseel (2015). blavaan: Bayesian Structural Equation Models via Parameter Expansion.

See Also

bcfa, bsem, bgrowth

Examples

dpriors(nu = "dunif(0,10)", lambda = "dnorm(0,1e-2) T(0,)", itheta = "dexp(1)")