www.cmm.bristol.ac.uk/forum

Hi, Thank you for your kind help! Actually I am very beginner of using MLwiN Commands as well as MCMC method. The one that I am using is Metropolis-Hastings algorithm with non-informative priors. So, manually (without using commands interface), I have run models first by running IGLS (automatically ...

Agreed with ChrisCharlton's suggestion of running MCMC. I also encountered similar issue for a four-level random intercept model with binary outcome. MCMC worked well while the 2nd order PQL had an issue of convergence.
Thanks,
YongJoo

Hello, I have just started to use MLwiN Macro. Mostly I am learning macro commands by checking what appears on the Command Interface window after I run a procedure by manually (clicking button by using mouse). It has worked well except for running models. For IGLS estimation, "STAR" comman...