coxme in R underestimates variance of random effect, when random effect is on observation level
I have a question concerning fitting a cox model with a random intercept, also known as a frailty model. I am using both the coxme package, and the frailty statement in coxph. Often 'shared' frailty models are implemented in practice, to group people who are from a cluster to account for homogeneity in outcomes for people from the same cluster. I am more interested in the classic frailty model, 'Early frailty models incorporated subject-specific random effects to account for unmeasured subject characteristics that influenced the hazard of the occurrence of the outcome'. This is because I have data where I would like to estimate the amount of variation between patients risks that has not been accounted for by adjusting for the variables that I have.