Thanks to John and Dale I can now use Proc Mixed to estimate the
parameters from my simulation.
During my search for information on mixed models I was told about these
which have examples from the text Applied Longitudinal Data Analysis:
Modeling Change and Event Occurrence (2003, Oxford).
The second site has worked examples in SAS, Stata, SPlus, SPSS, Mplus,
MLwiN and HLM.
This site (http://www.gllamm.org/) covers the Stata routine -gllamm-
(generalised linear latent and mixed models), which may be of interest