Issue with mplus2lavaan

[mhsteppan]

Amazing package! So practical!

I am trying to translate mplus example 5.1 to lavaan syntax. However, I get the following error message:

Error in sub("^([^@]+)(@[\d-.]+)?$", "\2", v_post, perl = TRUE) : invalid regular expression '^([^@]+)(@[\d-.]+)?$' In addition: Warning message: In sub("^([^@]+)(@[\d-.]+)?$", "\2", v_post, perl = TRUE) : PCRE pattern compilation error 'invalid range in character class' at '-.]+)?$'

I downloaded example 5.1 directly from the Mplus website: https://www.statmodel.com/usersguide/chap5/ex5.1.inp

Kind regards, Martin

[mjbsp]

I think there is a similar problem in the mplus2lavaan.modelSyntax.

The example code for this function gives me a similar error as @mhsteppan.

> syntax <- '
+     f1 BY x1*1 x2 x3;
+     x1 WITH x2;
+     x3 (1);
+     x2 (1);
+ '
>  mplus2lavaan.modelSyntax(syntax)
Error in gsub("(start\\([^\\)]+\\)\\*|[\\d-\\.]+\\*)", "", cmd.split,  : 
  invalid regular expression '(start\([^\)]+\)\*|[\d-\.]+\*)'
In addition: Warning message:
In gsub("(start\\([^\\)]+\\)\\*|[\\d-\\.]+\\*)", "", cmd.split,  :
  PCRE pattern compilation error
	'invalid range in character class'
	at '-\.]+\*)'

I did further digging into the code and I think that this particular error is in the function parseConstraints. In this function, I get the error at this line:

cmd.nomodifiers <- paste0(gsub("(start\\([^\\)]+\\)\\*|[\\d-\\.]+\\*)", "", cmd.split, perl=TRUE), collapse=" ") #peel off premultiplication

[yrosseel]

I have notified the maintainer of mplus2lavaan() (Michael Hallquist). Hopefully, he will find time to fix this soon.

[andrew0302]

Hello all: I'm also getting the same error, both with the sample file and other files.

Has anyone managed a a fix? Alterantively, is there a known older version that runs without this error?

[yrosseel]

It would seem my email (to Michael Hallquist) was never sent. Not sure why. Sorry for the delay.

[michaelhallquist]

Hi all, sorry this issue has been unresolved for some time. It was just a small oversight in the regular expression -- I hadn't escaped the hyphen used to match negative numbers. The PR above fixes this and I verified the results with Mplus 5.1 and the example syntax above.

[linem7]

Could Mplus2lavaan read multi-group analysis files？ I get the following error message: Error in gregexpr("(?!<(\\*|\\.))\\w+(?!(\\*|\\.))\\s*-\\s*(?!<(\\*|\\.))\\w+(?!(\\*|\\.))(@[\\d\\.-]+)?", :

My Mplus code is: `title: no days multigroup analysis; data: file is nodays.dat; vari: names are risk risk_cb12 ses gender age material space aut1 aut2 aut3 aut4 aur1 aur2 aur3 aur4 anx1 anx2 anx3 anx4 anx5; usev = risk_cb12 ses gender age material space aut1 aut2 aut3 aut4 aur1 aur2 aur3 aur4 anx1 anx2 anx3 anx4 anx5; grouping is risk_cb12(1 = low_medium 2 = high);

analysis: bootstrap = 10000;

model: inv by material space; aut by aut1 ant2 ant3 aut4; aur by aur1 aur2 aur3 aur4; anx by anx1 anx2 anx3 anx4 anx5;

inv on ses; aut on ses; aur on ses; anx on ses inv aut aur gender age; aut with aur;

model high:

inv on ses; aut on ses; aur on ses; anx on ses inv aut aur gender age; aut with aur;

output: cint(bcboot) stand;`

my R code is : mplus2lavaan("nodays2.inp", run = F)

thank you guys

[michaelhallquist]

Hi there, I don't currently plan to extend mplus2lavaan to support multi-group models. The syntax between Mplus and lavaan diverges quite a bit in this setting and I don't have time to work on this at present.