|
Message/Author |
|
|
Drs. Muthen, I received the following error message: 978 perturbed starting value run(s) did not converge or were rejected... 300 perturbed starting value run(s)... THE MODEL ESTIMATION DID NOT TERMINATE NORMALLY DUE TO AN INSUFFICIENT NUMBER OF E STEPS. INCREASE THE NUMBER OF MITERATIONS. ESTIMATES CANNOT BE TRUSTED. The DV and the indicators in the latent IV are moderately correlated (.35-.66). I used this model with a different DV and it ran just fine. Is this the problem or can you see something else being the problem? ANALYSIS: ESTIMATOR IS MLR; TYPE = COMPLEX MIXTURE; STARTS = 1000 300; PROCESSORS = 4; ITERATIONS = 20000; MITERATIONS = 20000; MODEL: %OVERALL% chrnctrl BY OACA_ART OACC_ARt ODC_ARt OVC_ARt; !covariance for identification OACC_ART WITH OACA_ART; chrnctrl ON CLINWRR C_PERWH (reg_gen); C_PERWH CLINWRR; %C#1% [chrnctrl] (m_1); chrnctrl OVC_ARt@.001; chrnctrl ON CLINWRR (reg1); %C#2% [chrnctrl] (m_2); chrnctrl@.001 OVC_ARt@.001; chrnctrl ON CLINWRR (reg2); %C#3% [chrnctrl] (m_3); chrnctrl OVC_ARt@.001; chrnctrl ON CLINWRR (reg3); %C#4% [chrnctrl] (m_4); chrnctrl OVC_ARt@.001; chrnctrl ON CLINWRR (reg4); |
|
Back to top |
|
|