regression: Add regression association to a latent variable model with...

In bozenne/lava.reduce: Latent Variable Models with linear predictor

Description

Same as regression for latent variable model with an argument reduce to introduce linear predictors

Usage

## S3 replacement method for class 'lvm.reduced'
regression(object = lvm.reduced(), ...) <- value

## S3 method for class 'lvm.reduced'
regression(object = lvm.reduced(), to, from, y, x,
  reduce = FALSE, value, ...)

Arguments

object	lvm-object
...	additional arguments to be passed to regression.lvm
value	A formula specifying the linear constraints or if to=NULL a list of parameter values.
to	Character vector of outcome(s) or formula object.
from	Character vector of predictor(s).
y	Alias for 'to'
x	Alias for 'from'
reduce	should the from variable be grouped into a linear predictor

Examples

m <- lvm.reduced()
m <- regression(m,y='y1',x='x'%++%1:2)
m <- regression(m,y='y1',x='z'%++%1:5, reduce = TRUE)
m

m <- lvm.reduced()
regression(m) <- y1 ~ x1 + x2
regression(m, reduce = TRUE) <- y1 ~ x5 + x6 + x7
regression(m, reduce = TRUE) <- as.formula(paste0("y~",paste("x",1:5,collapse="+",sep="")))

m <- lvm.reduced()
regression(m, reduce = "LL") <- y1 ~ x5 + x6 + x7
m

bozenne/lava.reduce documentation built on May 13, 2019, 1:41 a.m.