A Cox model for radioactive counting measure: Inference on the intensity process

A solution to the problem of calibrating a counting device from observed data, is developed in this paper by means of a Cox process model. The stochastic intensity of the process for counting emitted particles is estimated by functional principal components analysis and confidence bands are provided for two radioactive isotopes, 226Ra and 137Cs. A hypothesis test to assess the coherence of the new observed data with the estimated model is also included.