umxLav2RAM lavaanMode = "lavaan" weirdness?

[lf-araujo]

Dear Tim,

I came across a strange behaviour while trying to convert a lavaan model to RAM. In the model specification:

library(umx)
model <- "
  # random intercepts
  xir =~  1*X1 + 1*X2  + 1*X3 +1*X4
  yir =~  1*Y1 + 1*Y2 + 1*Y3 + 1*Y4

  # crosslaged
  Y2 ~ 0.1*X1
  X2 ~ 0.5*Y1
  Y3 ~ 0.1*X2
  X3 ~ 0.5*Y2
  Y4 ~ 0.1*X3
  X4 ~ 0.5*Y3

  # instrument
  X1 + X2 + X3 + X4  ~ 0.05*PGSx
  Y1 + Y2 + Y3 + Y4  ~ 0.1*PGSy

  # innovations
  innoX1 =~ 1*X1
  innoX2 =~ 1*X2
  innoX3 =~ 1*X3
  innoX4 =~1*X4
  innoY1 =~1*Y1
  innoY2 =~1*Y2
  innoY3 =~1*Y3
  innoY4 =~1*Y4

  X2  ~ 0.4*innoX1
  X3  ~ 0.3*innoX2
  X4 ~ 0.3*innoX3
  Y2  ~ 0.3*innoY1
  Y3  ~ 0.3*innoY2
  Y4 ~ 0.3*innoY3

  # correlations
  xir ~~ 0.5*yir

  #means
  X1 ~ 1
  X2 ~ 1
  X3 ~ 1
  X4 ~ 1
  Y1 ~ 1
  Y2 ~ 1
  Y3 ~ 1
  Y4 ~ 1

  # variances
  innoX1 ~~ 1*innoX1
  innoX2 ~~ 1*innoX2
  innoX3 ~~ 1*innoX3
  innoX4 ~~ 1*innoX4
  innoY1 ~~ 1*innoY1
  innoY2 ~~ 1*innoY2
  innoY3 ~~ 1*innoY3
  innoY4 ~~ 1*innoY4
"

Now when we look at the A matrix free values:

 m1 <- umxLav2RAM(model, lavaanMode = "lavaan", autoRun = F, printTab = F)
  manifests <- m1$manifestVars
  latents <- m1$latentVars
  allVariables <- c(manifests, latents)

  Afree <- matrix(m1@matrices$A@free, ncol = length(allVariables))

It is all set to FALSE, when in fact it should look more like the following, no?

> Afree
         Y2    X2    Y3    X3    Y4    X4    X1    Y1  PGSx  PGSy   xir   yir innoX1 innoX2 innoX3 innoX4 innoY1 innoY2 innoY3 innoY4
 [1,] FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE  TRUE FALSE FALSE  FALSE  FALSE  FALSE  FALSE   TRUE  FALSE  FALSE  FALSE
 [2,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE  TRUE FALSE FALSE FALSE   TRUE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE
 [3,] FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE  FALSE  FALSE  FALSE  FALSE  FALSE   TRUE  FALSE  FALSE
 [4,]  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE  FALSE   TRUE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE
 [5,] FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE   TRUE  FALSE
 [6,] FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE  FALSE  FALSE   TRUE  FALSE  FALSE  FALSE  FALSE  FALSE
 [7,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE
 [8,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE
 [9,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE
[10,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE
[11,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE
[12,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE
[13,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE
[14,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE
[15,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE
[16,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE
[17,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE
[18,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE
[19,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE
[20,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE  FALSE

With TRUE in the paths specified to be estimated.

• What am I doing wrong? Is it a bug?

[mcneale]

I think the Lavaan syntax you are using implicitly fixes parameters when, e.g., 1*X1 is used. Changing the code to the following at least freed up some of the elements of m1@matrices$A@free - which are correctly loaded into Afree. library(umx) model <- "

random intercepts

xir =~ X1 + X2 + X3 + X4 yir =~ 1Y1 + 1Y2 + 1Y3 + 1Y4

crosslaged

Y2 ~ 0.1X1 X2 ~ 0.5Y1 Y3 ~ 0.1X2 X3 ~ 0.5Y2 Y4 ~ 0.1X3 X4 ~ 0.5Y3

instrument

X1 + X2 + X3 + X4 ~ 0.05PGSx Y1 + Y2 + Y3 + Y4 ~ 0.1PGSy

innovations

innoX1 =~ 1X1 innoX2 =~ 1X2 innoX3 =~ 1X3 innoX4 =~1X4 innoY1 =~1Y1 innoY2 =~1Y2 innoY3 =~1Y3 innoY4 =~1Y4

X2 ~ innoX1 X3 ~ innoX2 X4 ~ innoX3 Y2 ~ innoY1 Y3 ~ innoY2 Y4 ~ innoY3

correlations

xir ~~ 0.5*yir

means

X1 ~ 1 X2 ~ 1 X3 ~ 1 X4 ~ 1 Y1 ~ 1 Y2 ~ 1 Y3 ~ 1 Y4 ~ 1

variances

innoX1 ~~ 1innoX1 innoX2 ~~ 1innoX2 innoX3 ~~ 1innoX3 innoX4 ~~ 1innoX4 innoY1 ~~ 1innoY1 innoY2 ~~ 1innoY2 innoY3 ~~ 1innoY3 innoY4 ~~ 1innoY4 "

And I note that code and /code (inside less-than and greater-than chevrons) gets weirdly formatted on github.

[lf-araujo]

Thanks, ok, not a bug. The lavaan switch is not compatible with setting starting values with the 0.0*var syntax.

Edit For future reference in case other users come across this. The syntax start(0.3)*var seems to set the starting value whilst keeping the parameter free to be estimated.