Motivated by challenging problems in Evolutionary Biology and science, we propose an exact non-parametric Bayesian approach for estimating the intensity function of inhomogenous point processes. Our method does not rely on large data augmentation or approximations of the likelihood function.We propose an exact MCMC sampler for posterior inference, test its performance on simulated data, and show that our method outperforms other exact and approximate methods. We demonstrate the utility of the approach in Cox processes of event and count data, and a particular coalescent process for cell lineage tracking experiments.
After the seminar we will go to Studenterbaren for a drink. Everybody is more than welcome to join.
