The method is effectively a combination of ICAR and Bayesian Phase Model - we tested different sampling regimes (size and spatial clustering) and the effects of calibration process using simulated archaeological dataset to determine model performance both in terms of accuracy and precision
