Good morning, as this is my first post, I apologize in advance if I am missing any information.
I am running a meta-analytic SEM using the metaSEM package under the following environmental conditions:
OpenMx version: 2.5.2 [GIT v2.5.2]
R version: R version 3.2.2 (2015-08-14)
Platform: x86_64-apple-darwin13.4.0
MacOS: 10.11.3
Default optimiser: SLSQP
Package: metaSEM
Type: Package
Version: 0.9.6
Date: 2015-10-31
License: GPL (>=2)
LazyLoad: yes
Using the function tssem2(), the results produce a status1 code of 5. A google search and brief perusal of your errors wiki informed me that this code either does not exist or now does exist and should not. I would be happy to provide any additional information that may be needed to assess this error, and I thank you for your time! The syntax used that produced this status1 code follows.
Sincerely,
Jamie
A1 <- matrix(c(0,0,0,0,0,0,".1*rri",".1*rwm",".1*rbg",".3*rdec",".3*rLing", 0,0,0,0,0,0,0,0,0,".3*dw",0, 0,0,0,0,0,0,0,0,0,".3*dt",".3*lt", 0,0,0,0,0,0,0,0,0,".3*dd",0, 0,0,0,0,0,0,0,0,0,0,".3*lv", 0,0,0,0,0,0,0,0,0,0,".3*ll", 0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0 ), ncol=11, nrow=11, byrow=TRUE) dimnames(A1)<- list(c("rc","f","t","d","voc","lc","ri","wm","bgk","DEC","LING"), c("rc","f","t","d","voc","lc","ri","wm","bgk","DEC","LING")) A1 S1 <- matrix(c( ".2*e1",0,0,0,0,0,0,0,0,0,0, 0,".2*e2",0,0,0,0,0,0,0,0,0, 0,0,".2*e3",0,0,0,0,0,0,0,0, 0,0,0,".2*e4",0,0,0,0,0,0,0, 0,0,0,0,".2*e5",0,0,0,0,0,0, 0,0,0,0,0,".2*e6",0,0,0,0,0, 0,0,0,0,0,0,".2*e7",0,0,0,0, 0,0,0,0,0,0,0,".2*e8",0,0,0, 0,0,0,0,0,0,0,0,".2*e9",0,0, 0,0,0,0,0,0,0,0,0,1,".3*cor", 0,0,0,0,0,0,0,0,0,".3*cor",1 ), ncol=11, nrow=11, byrow=TRUE) dimnames(S1)<- list(c("rc","f","t","d","voc","lc","ri","wm","bgk","DEC","LING"), c("rc","f","t","d","voc","lc","ri","wm","bgk","DEC","LING")) S1 F1<-create.Fmatrix(c(1,1,1,1,1,1,1,1,1,0,0), as.mxMatrix=FALSE) dimnames(F1)<- list(c("rc","f","t","d","voc","lc","ri","wm","bgk"), c("rc","f","t","d","voc","lc","ri","wm","bgk","DEC","LING")) F1 tssem2_rem<- tssem2(tssem_rem, Amatrix=A1, Smatrix=S1, Fmatrix=F1, model.name="TSSEM2") summary(tssem2_rem)
Printer-friendly version
As of version 2.3.1, OpenMx has used status code 5 to warn the user that the Hessian matrix (the matrix of second partial derivatives of the fitfunction w/r/t the free parameters) is not positive-definite at the solution. To put it another way, the fitfunction is not locally convex at the solution. This is a sign that the solution does not meet second-order conditions for being a local minimum of the fitfunction.
As for what to do about it in your case...I'm afraid I don't know metaSEM at all. You could ask in the metaSEM forum. Mike Cheung is quite responsive to questions posted there.
From the model specification in A1, ri, wm, bgk are independent variables. Since you are analyzing a correlation matrix, the variances of them must be fixed at 1 in S1.
Moreover, we usually allow the independent variables to be correlated. Therefore, ri, wm, bgk, DEC, and LING should all be correlated in S1.
Let's see if it helps.