st: Structural equation model

[holendro singh <holendrosingh@gmail.com>]

Dear all,

I am trying to assess the time invariance of a construct in a three
time point longitudinal study. I wrote the sem do command like

sem ( d11@1<-D06 _cons@c1) (d12 d13 d14 d15 <-D06) ///
    ( d21@1<-D08 _cons@c1) (d22 d23 d24 d25 <-D08) ///
    ( d31@1<-D10 _cons@c1) (d32 d33 d34 d35 <-D10), ///
	var(e.d11 e.d12 e.d13 e.d14 e.d15) ///
	var(e.d21 e.d22 e.d23 e.d24 e.d25) ///
	var(e.d31 e.d32 e.d33 e.d34 e.d35) ///
	means(D06 D08 D10) ///
	var(D06 D08 D10)

where D06, D08, D10 are the three constructs. In most of the analyses
I did I did not get the upper confidence limit for RMSEA. Whether my
do command is wrong. I am struggling with this. I got the output

		
Fit statistic	Value	Description
		
Likelihood ratio	
chi2_ms(87)	3574.504	model vs. saturated
p > chi2	0.000
chi2_bs(105)	13222.411	baseline vs. saturated
p > chi2	0.000
		
Population error	
RMSEA	0.127	Root mean squared error of approximation
90% CI, lower bound	0.000
upper bound	.
pclose	0.000	Probability RMSEA <= 0.05
		
Information criteria	
AIC	75351.287	Akaike's information criterion
BIC	75630.339	Bayesian information criterion
		
Baseline comparison	
CFI	0.734	Comparative fit index
TLI	0.679	Tucker-Lewis index
		
Size of residuals	
SRMR	0.081	Standardized root mean squared residual
CD	0.963	Coefficient of determination
		
Here the upper 95% for RMSEA is missing. Not only in this analysis in
most of the analyses I got this proble. Please help.

Regards




-- 
Holendro Singh Chungkham,
Post-Doctoral Fellow,
Stress Research Institute,
Stockholm University,
Frescati Hagvag 16A,
114 19 Stockholm
www.stressresearch.su.se
E-mail: holendro.chungkham@stressforskning.su.se
Phone: +468 5537 8913
Mob:     +467 2938 1386
Fax:      +468 5537 8900

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/