npb: Function to fit nonparametric Bayes shrinkage

In lvhoskovec/mmpack: Methods for multipollutant mixtures analysis

Description

Fits nonparametric Bayes shrinkage model for continuous response data using Markov chain Monte Carlo (MCMC) methods

Usage

npb(
  niter,
  nburn,
  X,
  Y,
  W,
  scaleY = FALSE,
  priors,
  interact = FALSE,
  intercept = TRUE,
  XWinteract = FALSE
)

Arguments

niter	number of iterations
nburn	number of burn-in iterations
X	n by p matrix of predictor data
Y	n by 1 vector of continuous response data
W	n by q matrix of covariate data
scaleY	logical; if TRUE response will be centered and scaled before model fit, default is FALSE
priors	list of prior hyperparameters, see package documentation for details
interact	logical; if TRUE (default) include all pairwise interactions of predictors, if FALSE include main effects only
intercept	logical; indicates if an overall intercept should be estimated with covariates
XWinteract	logical; indicates if X and W can interact

Value

an object of class "npb", which has the associated methods:

• print (i.e. print.npb )

• summary (i.e. summary.npb )

• predict (i.e. predict.npb )

a list with components

• X: predictor data matrix

• Y: response data vector

• Z: matrix of pairwise multiplicative interactions of predictor data (only if interact = TRUE)

• W: covariate data

• alpha: DP parameter for main effects estimates

• alpha.2: DP parameter for interactions estimates

• beta: unscaled main effect regression coefficient estimates

• zeta: unscaled interaction regression coefficient estimates

• mu: mu estimates

• mu.2: mu2 estimates

• phi2inv: phi2inv estimates

• phi2inv.2: phi2inv.2 estimates

• gamma: unscaled covariate regression coefficient estimates

• sig2inv: sig2inv estimates

• pip.beta: inclusion indicators for main effects

• pip.zeta: inclusion indicators for interactions

lvhoskovec/mmpack documentation built on Aug. 21, 2021, 4:05 p.m.