Hello, Stata users,
I am currently using STATA MP 18 to run stpm2 models. My dataset contains approximately 430,000 observations, with around 200,000 subjects and 2,500 failures. The total analysis time is about 1,000,000.
I used stpm2 with the tvc option and included about 14 covariates in the model. When I ran stpm2 using the same dataset but for a different category (using slightly fewer observations), the model took around 10 minutes to run. However, one specific line of code is now taking over two hours to execute.
I am wondering if this is a normal occurrence for such a dataset size or if there might be an issue with my code or settings causing this delay. Interestingly, the same code took less than 10 minutes two days ago, but the dataset then had approximately 10,000 fewer observations.
Have any of you experienced similar issues? I would appreciate any advice or insights on whether this runtime is expected or suggestions on how to optimize the process.
Thank you!
I am currently using STATA MP 18 to run stpm2 models. My dataset contains approximately 430,000 observations, with around 200,000 subjects and 2,500 failures. The total analysis time is about 1,000,000.
I used stpm2 with the tvc option and included about 14 covariates in the model. When I ran stpm2 using the same dataset but for a different category (using slightly fewer observations), the model took around 10 minutes to run. However, one specific line of code is now taking over two hours to execute.
I am wondering if this is a normal occurrence for such a dataset size or if there might be an issue with my code or settings causing this delay. Interestingly, the same code took less than 10 minutes two days ago, but the dataset then had approximately 10,000 fewer observations.
Have any of you experienced similar issues? I would appreciate any advice or insights on whether this runtime is expected or suggestions on how to optimize the process.
Thank you!
