Generating survival times to simulate Cox proportional hazards models

Abstract

SUMMARYThis paper discusses techniques to generate survival times for simulation studies regarding Cox proportional hazards models. In linear regression models, the response variable is directly connected with the considered covariates, the regression coefficients and the simulated random errors. Thus, the response variable can be generated from the regression function, once the regression coefficients and the error distribution are specified. However, in the Cox model, which is formulated via the hazard function, the effect of the covariates have to be translated from the hazards to the survival times, because the usual software packages for estimation of Cox models require the individual survival time data. A general formula describing the relation between the hazard and the corresponding survival time of the Cox model is derived. It is shown how the exponential, the Weibull and the Gompertz distribution can be used to generate appropriate survival times for simulation studies. Additionally, the general relation between hazard and survival time can be used to develop own distributions for special situations and to handle flexibly parameterized proportional hazards models. The use of other distributions than the exponential distribution only is indispensable to investigate the characteristics of the Cox proportional hazards model, especially in non-standard situations, where the partial likelihood depends on the baseline hazard.

Sensitivity analysis for the effects of multiple unmeasured confounders

“…Our approach can be likewise extended to time-to-event data, in which case time-to-event (or censoring), as well as the outcome status would need to be simulated. 26 In our simulations, we evaluated unmeasured confounding as a possible reason for the discrepancy between randomized and nonrandomized studies regarding the effect of ascorbic acid on mortality. Other reasons for this discrepancy include (differential) misclassification of the exposure or confounders, selective loss to follow-up, missing data and different durations of ascorbic acid exposure.…”

A comparison of inverse transform and composition methods of data simulation from the Lindley distribution

“…Therefore, a numerical technique, such as the Newton-Raphson, may be employed to solve for the Lindley variates (Bender et al, 2005;Okwuokenye, 2012;Okwuokenye andPeace, 2015a, 2015b;Peace, 1976;Peace and Flora, 1978;Qian et al, 2010;Wicklin, 2013). We generated data from the Lindley distribution using inverse transform by iteratively solving…”

Resampling to Address the Winner's Curse in Genetic Association Analysis of Time to Event

“…The baseline cumulative hazard was estimated for each stratum using observed time to event for 1361 individuals from the original EDIC/DCCT dataset. The event-time variable for each individual was computed using an equation equivalent to Bender’s equation 6 [2005], and compared to a standard uniform variate. Where the uniform variate fell below the last step of the estimated survival curve, the event time was set to censored at the last visit.…”