Transformations, where the transformed parameter is not used in the model, do not work

[jhorzek]

The following does not work:

library(lessSEM)

# Identical to regsem, lessSEM builds on the lavaan
# package for model specification. The first step
# therefore is to implement the model in lavaan.

dataset <- simulateExampleData()

lavaanSyntax <- "
f =~ l1*y1 + l2*y2 + l3*y3 + l4*y4 + l5*y5 + 
     l6*y6 + l7*y7 + l8*y8 + l9*y9 + l10*y10 + 
     l11*y11 + l12*y12 + l13*y13 + l14*y14 + l15*y15
f ~~ 1*f
"

lavaanModel <- lavaan::sem(lavaanSyntax,
                           data = dataset,
                           meanstructure = TRUE,
                           std.lv = TRUE)

# Regularization:

lsem <- lasso(
  # pass the fitted lavaan model
  lavaanModel = lavaanModel,
  # names of the regularized parameters:
  regularized = paste0("l", 6:15),
  # in case of lasso and adaptive lasso, we can specify the number of lambda
  # values to use. lessSEM will automatically find lambda_max and fit
  # models for nLambda values between 0 and lambda_max. For the other
  # penalty functions, lambdas must be specified explicitly
  nLambdas = 50)

# use the plot-function to plot the regularized parameters:
plot(lsem)

transformation <- "
parameters: l3, l4, lprod
lprod= l3 * l4;
"

lsem <- lasso(
  # pass the fitted lavaan model
  lavaanModel = lavaanModel,
  # names of the regularized parameters:
  regularized = paste0("l", 6:15),
  # in case of lasso and adaptive lasso, we can specify the number of lambda
  # values to use. lessSEM will automatically find lambda_max and fit
  # models for nLambda values between 0 and lambda_max. For the other
  # penalty functions, lambdas must be specified explicitly
  lambdas = seq(0,1,.1),
  modifyModel = modifyModel(transformations = transformation))
coef(lsem)
lsem@transformations

Such transformations could be useful for path tracing rules, but we cannot regularize lprod anyway, so I will not implement these transformations for the time being. We could add a warning to let users know.

[jhorzek]

Maybe we could check if something that the user puts in as a parameter (here: lprod) gets retuned with a different value by the transformation function. That is, lessSEM expects that all SEM parameters (e.g., l1, l2) can be a function of other parameters. However, non-SEM parameters (lprod) are not allowed to be functions of other parameters.