Merlin syntax: Mopt_maxmin() error STATA17

[Mike Stark]

Dear Stata users,

I'm trying to run the code following this article "merlin—A unified modeling framework for data analysis and methods development in Stata". However, in the final model, I have faced this issue, when I set the trace on, it seems that because the Mopt_maxmin() mata is blank. Could you help me to solve this problem? I'm really appreciated for your time. thank you very much.

[Nick Cox]

The author Michael Crowther might or might not see this given a ping. If not, I would email him directly.

[Mike Stark]

The author Michael Crowther might or might not see this given a ping. If not, I would email him directly.

Dear Nick Cox,

Thank you very much for your kind support.

Regards,

[Michael Crowther]

Thanks for the post - this'll be fixed in the next update which should go up this week.

[Sharon Ayayo]

Hi,

I’m trying to fit transition-specific Cox models using the stmerlin command and returned the error below:

merlin_setup_core(): 3259 nonstruct found where struct required
merlin_setup(): - function returned error
<istmt>: - function returned error

I’ve tried to debug it but can’t see anything wrong with my command or variables.

I started by setting the data for multistate modelling as follows:
msset, id(pseudoID) states(bleeding_indic reinfarction_indic death_indic) times(time_bleeding time_reinfarction time_death) covariates(age sex ethnicity prevAMI hyperten hypercholes peripheral cerebrovasc renal smoking diabetic prevPCI prevCABG prevAngina famhist_chd p2y12 asp)

Then set it for survival analysis:
stset _stop, enter(_start) failure(_status=1) scale(365.25)

Then fitted the Cox model as follows:
ssc install merlin
ssc install moremata
ssc install gtools

stmerlin age i.sex i.ethnicity i.prevAMI i.hyperten i.hypercholes i.peripheral i.cerebrovasc i.renal i.smoking i.diabetic i.prevPCI i.prevCABG i.prevAngina i.famhist_chd i.p2y12 i.asp if _trans1==1, distribution(cox)

Seems the problem is with the structure of the variables, I tried running each variable at a time but none worked. Also I've imputed missing values in these variables, so I don't know where the problem is.

Your help in debugging this will be highly appreciated, thanks!

[Nick Cox]

See #2 again. Best to contact Michael directly.

[Bjarte Aagnes]

test again after updating from github:

net install merlin, from("https://raw.githubusercontent.com/RedDoorAnalytics/merlin/main/") replace
net install stmerlin, from("https://raw.githubusercontent.com/RedDoorAnalytics/stmerlin/main/") replace