search

drizopoulos / JMbayes2

Extended Joint Models for Longitudinal and Survival Data
https://drizopoulos.github.io/JMbayes2/
78 stars 22 forks source link

Confusing point estimates for association parameters in current values + current slopes model #54

Closed amirabadiza921 closed 1 year ago

amirabadiza921 commented 1 year ago

Hi Dr. Rizopoulos, When I run the current values+ current slopes, I get point estimates of assoct = 4.27 (SE = 1.03) and assoct.s = 28.10 (SE=4.86).

model.cox<-coxph(Surv(surv_time,incidenceCKD)~sex15_1+age+FHD+ waist+fbs+bs2HR+BMI,data=data.cox,x=T,model=TRUE) fit.lm1<-lme(FT4~time,random= ~time | code,data=dataLong) ffm<-list("FT4" = ~ value(FT4) + slope(FT4)) fit.JM<-jm(model.cox,fit.lm1,functional_forms = ffm,time_var="time") summary(fit.JM) Thanks

drizopoulos commented 1 year ago

It is not clear what exactly the issue is. If the model has converged, the magnitude of the coefficients needs to be interpreted, accounting for the scale of the longitudinal outcome. For example, is the random slopes are very small in magnitude per subject, it is logical to get a large coefficient for the slope.