tbates / umx

Making Structural Equation Modeling (SEM) in R quick & powerful
https://tbates.github.io/
44 stars 17 forks source link

v.m. too cryptic? #99

Closed tbates closed 4 years ago

tbates commented 4 years ago

Hi #1 user :-) The goal is less typing (1-liner to create free var and mean) while retaining clarity.

If v.m. fails, any thoughts for better synonym?

Trouble with "plain english" is that even "freeVarianceAndMean" is obscure to the person who thinks "MeansAndVarsFree". Might be no shortcut to some learning curve. And clarity seekers best to use the non-cryptic

umxPath(var = "bob")
umxPath(mean = "bob")

The only other thought I have is two new functions?

manifestVars("bob") # with default smart-starts for mean and variance
latentVars("jane") # with defaults varFixedAt = 1, meanFixedAt = 0

For background, most keywords are English, e.g. var = and means=

mcneale commented 4 years ago

Maybe there should be a umxGlossary()?

tbates commented 4 years ago

:-) maybe

umxGuess("residuals")

Perhaps you want umx_residualize?

I was just trying to understand an Mplus model (never can remember their syntax) and tried the mplus2lavaan.modelSyntax function, which didn't quite do it, but then running it into umxRAM() to get a plot... Kind of got there in the end.