pwSEM (version 1.0.0)

get.AIC: Title get.AIC

Description

Title get.AIC

Usage

get.AIC(sem.model, MAG, data)

Value

A data frame containing the log-likelihood of the full SEM (LL), the number of free parameters that were estimated (K), along with the AIC and the bias-corrected AIC (AICc)

Arguments

sem.model

A list containing the structural equations, each created using either the gam or the gamm functions of the mgcv package

MAG

A matrix encoding the directed acyclic graph (DAG), or the mixed acyclic graph, of the structural equations model. This is created using the DAG or the makeGM function in the gmm library

data

A data frame holding the observed data used in the calls to the models in the sem.model object

Examples

library(mgcv)
library(ggm)
set.seed(10)
N<-1000
L1<-rnorm(N)
x1<-0.5*L1+rnorm(N,0,sqrt(1-0.5^2))
x2<-0.5*x1+rnorm(N,0,sqrt(1-0.5^2))
x3<-0.5*L1+0.5*x2+rnorm(N,0,sqrt(1-2*0.5^2))
x4<-0.5*x3+rnorm(N,0,sqrt(1-0.5^2))
my.dat<-data.frame(x1,x2,x3,x4)

my.list<-list(gam(x1~1,data=my.dat),
             gam(x2~x1,data=my.dat),
             gam(x3~x2,data=my.dat),
             gam(x4~x3,data=my.dat))

full.dag<-DAG(x1~L1,x2~x1,x3~x2+L1,x4~x3)
my.mag<-DAG.to.MAG.in.pwSEM(full.DAG=full.dag,latents=c("L1"))
get.AIC(sem.model=my.list,MAG=my.mag,data=my.dat)