152
(Semi)parametric Example
Survival cure model
(Ibrahim, 1999, 2000)
–Probability model
•Proportion πi is
cured (survival probability 1 at ∞) in the i-th
treatment group
•Noncured group has
survival distribution modeled parametrically (e.g.,
Weibull) or semiparametrically (e.g., proportional hazards)
•Treatment effect is
measured by θ = π1 – π0
–The problem as I see
it: Incorrect assumptions about the nuisance parameter can bias
the estimation of the treatment effect