ufworthy.blogg.se

They can be created using the same vpc_tte() function by specifying the relevant predictor to the kmmc argument, as shown below. KMMC plots have been presented by Andy Hooker as a model diagnostic for covariate inclusion. The bins are basically based on all unique timepoints for all events the simulations dataset. It is therefore advised to not apply binning for time-to-event data, and to just use the 'bins' generated from the simulated data. However, the binning and smoothing often makes the bins go over the 0 - 100% interval and some elements of the confidence interval may not be shown correctly or even missing from the plot. In principle, the vpc_tte() function does allow binning and smoothing like commonly applied to a regular vpc. Obs_cols = list(idv = "t"), verbose=TRUE) Rtte = TRUE, rtte_calc_diff = TRUE, events = c(1:3),

Stratified for event number (RTTE) and study arm: vpc_tte(sim = rtte_sim_nm, Stratified for covariate and study arm, and binned and smooth: vpc_tte(sim = rtte_sim_nm, Treat RTTE as TTE, no stratification: vpc_tte(sim = rtte_sim_nm,

The datasets are supplied with the vpc library. Example will follow soon.Īn example for time-to-event data is shown below. This approach is straightforward in R but cannot easily be implemented in NONMEM. The disadvantage with this approach is however that the hazard is assumed constant over time, so models with time-dependent hazards cannot easily be simulated with this approach. Advantages of this approach is that it is much faster, and it does not require a dense grid.

one that has observation times at every possible timepoint that an event can occur for all individuals.ĭirect sampling: Sample event times directly from the distribution used to model the data (e.g. Also, a dataset with a dense design grid has to be used for simulation, i.e. The disadvantage of this method is that it is slow due to the numerical solving of the ODEs. Hazard integration: Integrate the hazard over time, and at any possible observation timepoint randomly draw a binary value based on the probability of observing the event. In general, there are two distinct approach to simulate survival data: Similar to the VPC for continuous data, the VPC for TTE data requires simulated data.