mxMmodModel {mxmmod}

Create an MMOD

Description

This function builds a Measurement Model of Derivatives (MMOD; Estabrook 2015) with a given factor structure.

Usage

mxMmodModel(
  data,
  modelName,
  idvar,
  timevar,
  structure,
  orthogonal = F,
  embed_dim = NULL,
  fiml = F
)

Arguments

data

a data frame with measurements in long format

modelName

name for the resulting model

idvar

name of column for subject IDs

timevar

name of column for measurement occasion

structure

factor structure, see 'Details'

orthogonal

if true, fix correlations between factors to 0 (A factor and its derivatives will still intercorrelate)

embed_dim

time delay embedding dimension

fiml

if true, use raw data to fit model with FIML. Otherwise, fit using cov matrix (dropping missing values if necessary).

Details

The structure argument is a list of latent factors and their mappings to manifest variables. For example, a one factor structure would be:

list(F1 = c('m1', 'm2', 'm3', 'm4', 'm5', 'm6'))

And a two factor structure would be:

list(F1 = c('m1', 'm2', 'm3'), F2 = c('m4', 'm5', 'm6'))

Value

an MMOD as an mxModel object

Examples

data(nlsy97depression)
# Fit one factor MMOD
structure <- list(
  F1 = c('nervous', 'down', 'depressed', 'calm', 'happy')
)
mmod_model <- mxMmodModel(data=nlsy97depression,
                          modelName='1 Factor MMOD',
                          idvar='pid', timevar='occasion', structure=structure)
mmod_fit <- OpenMx::mxRun(mmod_model)
summary(mmod_fit)