A Gaussian copula joint model for longitudinal and time-to-event data with random effects

Longitudinal and survival sub-models are two building blocks for joint modelling of longitudinal time-to-event data. Extensive research indicates separate analysis these processes could result in biased outputs due to their associations. Conditional independence between measurements biomarkers event time process given latent classes or random effects is a conventional approach characterising the association while taking heterogeneity among population into account. However, this assumption tricky validate because unobservable variables. Thus Gaussian copula model with proposed accommodate scenarios where conditional questionable. The assuming special case when parameters shrink zero. Simulation studies real data application carried out evaluate performance different correlation structures. In addition, personalised dynamic predictions probabilities obtained based on comparisons made under model.