Figure 2

Effect of risk-free intervals on cumulative hazard estimation. Pointwise mean cumulative hazard estimates (solid lines) and corresponding 2.5% and 97.5% quantiles derived from 1000 simulated datasets with 100 observations each, that are observed over the follow-up period [0,2]. Data are generated without risk-free intervals (continuous risk intervals) and with risk-free intervals of length d=3/12 following each event (discontinuous risk intervals), respectively. Data distribution is Weibull with \(\text {scale}=4/\sqrt {2}\) and shape=0.5. Mean estimates derived from the two simulation models are overlapping.