Easy question regarding SEM in lisrel

#1
Hey all,

I'm running a full SEM in lisrel and I'm having an issue. Below is the code. The issue is namely this: I specified Lambda Y as full and fixed, and then freed my 6 relevant parameters (1,1; 2,1; 3,2; 4,2; and 5;3). HOWEVER, Lisrel is only estimating LY(2,1) and LY(4,2) and the rest appear on my path diagram as numbers, but the lines are grey and there is no error on the table (as well in the initial output it is clear Lisrel is only allowing those two elements of LY to be free). Does anyone have any idea why this is happening and how to fix it?


Ideology and Control Full Model V1
DA NI=11 NO=393 MA=CM
CM SY
*
1.064
.229 .963
-.060 -.117 .816
.088 -.140 .562 1.031
.397 .097 .373 .526 1.769
.134 .032 .005 .132 .183 1.066
-.051 -.006 .367 .587 .494 .182 1.949
.023 -.014 .238 .414 .017 .052 .603 3.989
.750 .109 .285 .479 .703 .196 .402 .143 2.348
.142 .149 -.350 -.427 -.098 -.015 -.204 -.258 -.070 .924
.031 .314 -.339 -.287 -.101 .156 .161 .070 -.084 .619 1.968
LA
*
ECON.ATT LIB SOCIAL.ATT RWA ID UA RTC TM SDO LOC PC
MO NX=6 NY=5 NK=4 NE=3 BE=FU,FI GA=FU,FI PH=FU,FR PS=DI,FR LX=FU,FI LY=FU,FI TD=DI,FI TE=DI,FI
FR BE(3,1) BE(3,2) BE(1,2) BE(2,1)
FR GA(1,4) GA(2,1) GA(2,2) GA(2,3) GA(2,4)
FR LX(1,1) LX(2,1) LX(3,2) LX(4,3) LX(5,4) LX(6,4)
FR LY(1,1) LY(2,1) LY(3,2) LY(4,2) LY(5,3)
FR TD(1,1) TD(2,2) TD(5,5) TD(6,6) TE(1,1) TE(2,2) TE(3,3) TE(4,4)
VA .85 TD(3,3) TD(4,4) TE(5,5)
PD
OU SE TV ME=ML EF RS ND=6 SS MI
 

Lazar

Phineas Packard
#2
I find reading lisrel syntax near on impossible having not touched it in many years. Can you post the model implied matrices at the beginning of the output?
 
#3
Absolutely! here they are. Thanks!

Ideology and Control Full Model V1

Parameter Specifications

LAMBDA-Y
ETA 1 ETA 2 ETA 3
-------- -------- --------
ECON.ATT 0 0 0

LIB 1 0 0

SOCIAL.A 0 0 0

RWA 0 2 0

ID 0 0 0

LAMBDA-X

KSI 1 KSI 2 KSI 3 KSI 4

-------- -------- -------- --------

UA 3 0 0 0

RTC 4 0 0 0

TM 0 5 0 0

SDO 0 0 6 0

LOC 0 0 0 7

PC 0 0 0 8

BETA

ETA 1 ETA 2 ETA 3

-------- -------- --------

ETA 1 0 9 0

ETA 2 10 0 0

ETA 3 11 12 0

GAMMA

KSI 1 KSI 2 KSI 3 KSI 4

-------- -------- -------- --------

ETA 1 0 0 0 13

ETA 2 14 15 16 17

ETA 3 0 0 0 0

PHI

KSI 1 KSI 2 KSI 3 KSI 4

-------- -------- -------- --------

KSI 1 0

KSI 2 18 0

KSI 3 19 20 0

KSI 4 21 22 23 0

PSI

ETA 1 ETA 2 ETA 3

-------- -------- --------

24 25 26

THETA-EPS

ECON.ATT LIB SOCIAL.A RWA ID

-------- -------- -------- -------- --------

27 28 29 30 0

THETA-DELTA

UA RTC TM SDO LOC PC

-------- -------- -------- -------- -------- --------

31 32 0 0 33 34
 
#5
In the syntax I have linked I know the model is unidentified, however unless I am doing something wrong when I run it restricted BE(2,1) and BE(1,2) the model should be identified but Lisrel does the same thing.