Abstract
In this paper we generalise the risk models beyond the ordinary framework of affine processes or Markov processes and study a risk process where the claim arrivals are driven by a Cox process with renewal shot-noise intensity. The upper bounds of the finite-horizon and infinite-horizon ruin probabilities are investigated and an efficient and exact Monte Carlo simulation algorithm for this new process is developed. A more efficient estimation method for the infinite-horizon ruin probability based on importance sampling via a suitable change of probability measure is also provided; illustrative numerical examples are also provided.
Original language | English |
---|---|
Pages (from-to) | 55-65 |
Number of pages | 11 |
Journal | Insurance: Mathematics and Economics |
Volume | 65 |
DOIs | |
Publication status | Published - 1 Nov 2015 |