topic Time-dependent cox model in Discussions

Time-dependent cox model

cjw99 — Wed, 09 Oct 2024 21:25:35 GMT

I have been a loyal JMP user for more than ten years, and I have introduced many of my colleagues and students to JMP. I love its user interface and the speed with which I can do analysis.

However, more and more clinical journals in my field (transplantation) require time-dependent models (usually Cox models) for publication (see Gaynor et al. Am J Transplant. 2023 p.815 recently highlighting this in the number one transplant journal in the world). Thus, as a clinician, I am increasingly forced to rely on my statistical colleagues using R to help me conduct this analysis since JMP lacks this ability.

This is extremely frustrating since I can perform all other univariable and multivariate analyses myself, but as soon as I hit this roadblock, it slows down my workflow because I now have to depend on others to do my analysis.

Although I would love to keep using JMP, at this point, it is becoming harder and harder not to bite the bullet and learn how to use R (I really don’t want to spend the time learning how to code).

Can you tell me if JMP ever plans to support time-dependent Cox models?

Re: Time-dependent cox model

Thierry_S — Thu, 10 Oct 2024 01:12:04 GMT

Hi,

I might be naive, but would the Proportional Hazard analysis in the Survival and Reliability platform meet your requirements?

https://www.jmp.com/support/help/en/17.2/?os=win&source=application#page/jmp/fit-proportional-hazards.shtml#

Best regards,

Re: Time-dependent cox model

Mark_Bailey — Thu, 10 Oct 2024 13:24:19 GMT

You can use R with JMP. Your colleagues could write the R script for you. You will need a little JMP script to run it. You can send data and variables to R and return results from R for further analysis and display in JMP.