Vignette output mismatch and high sensitivity to random seeds

[valentinecleach]

Hello,
Thank you for developing and maintaining this package!
I've been working through the "Worked Example: Complete Analysis" vignette and have a couple of questions about consistency and reproducibility that I'd appreciate some clarity on.
• While running the vignette with the seed 2026, I get different results than the vignette for the relaxed model. I am however finding the same parameters for the SPFA model. Any idea what could cause this difference?
• When I change the seed used to generate my data, I see large shifts in the parameter estimates. These differences seem larger what could be explained by the differences in patient characteristics. Is this expected behavior, or could it point to an identifiability issue?
Any insight into either point would be greatly appreciated. Thanks in advance for your time and help!
Valentine