Learn R Programming

semPlot (version 1.1.6)

cvregsem: Bridge between cv_regsem output and sempaths

Description

The package regsem (Jacobucci, 2017) is designed for a specific type of SEM called regularized structural equation modelling (RegSEM). For more information about RegSEM and the implementation in R we refer to the manual written by Jacobucci (2017).This function creates a bridge between the regsem and semplot packages, making it possible to use output from the regsem() and cv_regsem() functions to create models in sempaths.

Usage

# S3 method for cvregsem
semPlotModel(object,model,...)

Value

A 'semPlotModel' object.

Arguments

object

The regsem output

model

The cfa output used as input for the cv_regsem function

...

Arguments sent to 'lisrelModel', not used in other methods.

Author

Sacha Epskamp <mail@sachaepskamp.com> Jason Nak <jasonnak@hotmail.com> Myrthe Veenman <myrthe.veenman@hotmail.com>

References

Jacobucci, R. (2017). regsem: Regularized Structural Equation Modeling. arXiv preprint arXiv:1703.08489.

See Also

semPlotModel semPaths

Examples

Run this code
## Example of fitting and plotting a cv_regsem model in semPaths

#library(psych)
#library(lavaan)
#library(regsem)

# use a subset of the BFI
#bfi2 <- bfi[1:250,c(1:5,18,22)]
#bfi2[,1] <- reverse.code(-1,bfi2[,1])

# specify a SEM model
#mod <- "
#f1 =~ NA*A1+A2+A3+A4+A5+O2+N3
#f1~~1*f1
#"

# fit the model
#fit <- cfa(mod, bfi2)
#out.reg <- cv_regsem(fit, type="lasso", pars_pen=c(1:7), n.lambda=23, jump =.05)

# plot the model
#semPaths(semPlotModel.cvregsemplot(object = out.reg, model = fit))

Run the code above in your browser using DataLab