Skip to main content
Figure 3 | BMC Medical Research Methodology

Figure 3

From: Simulating recurrent event data with hazard functions defined on a total time scale

Figure 3

Regression parameter estimation of Andersen-Gill model in the presence of unadjusted inter-patient heterogeneity. Mean regression coefficient estimates with mean 95% confidence interval limits as derived from Andersen-Gill analysis applying naive and robust standard errors, respectively. Results are derived from 1000 simulated datasets each reflecting recurrent event data from 100 individuals with binary covariate X and regression coefficient β=1 and with gamma-distributed frailty term with mean 1 and variance θ, that reflects inter-patient heterogeneity. Baseline hazard is defined as Weibull with \(\text {scale}=4/\sqrt {2}\) and shape=0.5.

Back to article page