drizopoulos
commented
3 years ago Dear Ben,
The standard version of joint models, but also of time-varying Cox models assumes that the association coefficient is the same for all time points. That is, the biomarker is time-varying, the coefficients for the biomarker not. There are version that relax this, i.e., to fit a model with a time-varying coefficient.
Best, Dimitris
From: Ben notifications@github.com Sent: Thursday, January 28, 2021 3:06 PM To: drizopoulos/JMbayes JMbayes@noreply.github.com Cc: Subscribed subscribed@noreply.github.com Subject: [drizopoulos/JMbayes] Extract JM estimated value and slope at time t (#83)
Dear prof,
What is the best way to assess the JM assumed value and slope of my longitudinal outcome at a certain time t? E.g. patients are followed from day 0 to 50 and measured. I fit a JMbayes value and slope model. How would I acquire the JM estimated value (predict()?) and slope at t=e.g. 25 or at time 125 (if there is a difference)?
Best,
Ben
- You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fdrizopoulos%2FJMbayes%2Fissues%2F83&data=04%7C01%7Cd.rizopoulos%40erasmusmc.nl%7Cbc6df3653c30456bd61008d8c395c6cf%7C526638ba6af34b0fa532a1a511f4ac80%7C0%7C0%7C637474395340457562%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=DCMbBwR15Co4YiU07TJWBGW2F1s8MFtxRJgCFWETH4E%3D&reserved=0, or unsubscribehttps://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FADE7TT3CBG4QODBAOZH3XP3S4FVKXANCNFSM4WXDU2NQ&data=04%7C01%7Cd.rizopoulos%40erasmusmc.nl%7Cbc6df3653c30456bd61008d8c395c6cf%7C526638ba6af34b0fa532a1a511f4ac80%7C0%7C0%7C637474395340457562%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=hrrlI4AlJMS4NAxVHiQvExrHXVjPkgDRQN2XXypUZ%2FI%3D&reserved=0.
Dear prof,
What is the best way to assess the JM assumed value and slope of my longitudinal outcome at a certain time t? E.g. patients are followed from day 0 to 50 and measured. I fit a JMbayes value and slope model. How would I acquire the JM estimated value (predict()?) and slope at t=e.g. 25 or at time 125 (if there is a difference)?
Best,
Ben