Help with interpreting coefficient of determination in SEM output

[Al Bothwell]

Hi Statalist,

I ran a CFA using SEM in Stata 18.5.
I obtained a "convergence not achieved" message using this code:

sem (F1 -> a1 a2 a3 a4 a5 a5 a6 a7 a8 a9) (F2 -> b1 b2 b3 b4 b5 b6 b7 b8) (F3 -> c1 c2 c3 c4 c5) (F4 -> d1 d2 d3), ///
vce(oim) method(ml) latent(F1 F2 F3 F4 ) nocapslatent standardized

So I changed the method from maximum likelhood to ADF—asymptotic distribution free. The model run successfully but the coefficient of determination is 1, which I understand indicates perfect fit. I always get nervous when a statistic reports a perfect fit. How do I explain this finding? I am doing something wrong?

sem (F1 -> a1 a2 a3 a4 a5 a5 a6 a7 a8 a9) (F2 -> b1 b2 b3 b4 b5 b6 b7 b8) (F3 -> c1 c2 c3 c4 c5) (F4 -> d1 d2 d3), ///
vce(bootstrap) method(adf) latent(F1 F2 F3 F4 ) nocapslatent standardized
estat eqgof
estat gof, stats(all)

Regards,
Al Bothwell