psspatmodelglm: Spatial propensity score model for Poisson outcomes.

In czigler/HEIfunctions: Functions for HEI Air pollution accountability grant

Description

Spatial propensity score model for Poisson outcomes.

Usage

psspatmodelglm(formula, data, trt, coords, family = "poisson", na.rm = T, nsamp, nburn, tau.sq.IG, sigma.sq.IG, tau.sq.start = 1, sigma.sq.start = 1, w.start = 0, phi.tune = 0.01, sigma.sq.tune = 0.15, tau.sq.tune = 0.15, beta.tune = rep(0.1, 5), w.tune = 0.1, phiprior = NULL)

Arguments

formula	formula
data	dataset
trt	string variable with name of intervention covariate, coded as 0 or 1.
coords	coordinates corresponding to data
family	right now, only works with family = "poisson"
na.rm	if na.rm=T, outcomes with missingness are deleted, and only counterfactuals are imputed. if = F, missing outcomes are imputed.
nsamp	total number of samples
nburn	number of samples to discard as burnin
tau.sq.IG	prior for tau.sq
sigma.sq.IG	prior for sigma.sq
tau.sq.start	starting value for tau.sq
sigma.sq.start	starting value for sigma.sq
w.start	starting value for w
phi.tune	tuning parameter for phi
sigma.sq.tune	tuning parameter for sigma.sq
tau.sq.tune	tuning parameter for tau.sq
beta.tune	vector of tuning parameters for beta
w.tune	tuning parameter for w
phiprior	prior for phi

Author(s)

Lauren

czigler/HEIfunctions documentation built on May 14, 2019, 1:43 p.m.