impacts.SEMCMC.xxx: Compute impacts for different models

In becarioprecario/SEMCMC: Bayesian Spatial Econometrics Models with MCMC using Jags and Stan

Description

This is an internal function to compute the impacts (direct, indirect and total) from a SEMCMC object.

Usage

## S3 method for class 'SEMCMC.sem'
impacts(obj, idx.var, var.names)

## S3 method for class 'SEMCMC.slm'
impacts(obj, W, idx.var, var.names)

## S3 method for class 'SEMCMC.sdm'
impacts(obj, W, idx.var, var.names)

## S3 method for class 'SEMCMC.sdem'
impacts(obj, W, idx.var, var.names)

Arguments

obj	A SEMCMC object.
idx.var	An index to subset the covariates coeffiecients
var.names	Vector with variable names
W	An adjacency matrix, same as used in the call to SEMCMC().

Examples

data(columbus, package = "spdep")

W <- spdep::nb2mat(col.gal.nb, style = "W")
m.form <-  CRIME ~ INC + HOVAL
#SEM model
sem.mcmc <- SEMCMC(m.form, data = columbus, W = W, model = "sem")
impacts(sem.mcmc, W)
data(columbus, package = "spdep")

W <- spdep::nb2mat(col.gal.nb, style = "W")
m.form <-  CRIME ~ INC + HOVAL
#SLM model
slm.mcmc <- SEMCMC(m.form, data = columbus, W = W, model = "slm")
impacts(slm.mcmc, W)
data(columbus, package = "spdep")

W <- spdep::nb2mat(col.gal.nb, style = "W")
m.form <-  CRIME ~ INC + HOVAL
#SDM model
sdm.mcmc <- SEMCMC(m.form, data = columbus, W = W, model = "sdm")
impacts(sdm.mcmc, W)
data(columbus, package = "spdep")

W <- spdep::nb2mat(col.gal.nb, style = "W")
m.form <-  CRIME ~ INC + HOVAL
#SDEM model
sdem.mcmc <- SEMCMC(m.form, data = columbus, W = W, model = "sdem")
impacts(sdem.mcmc, W)

becarioprecario/SEMCMC documentation built on Feb. 25, 2021, 4:03 p.m.