Simulating perpetuities
Perpetuity is the random variable \(Y=1+W_1+W_1W_2+W_1W_2W_3+\dots\), where \(W_i\) are independent nonnegative identically distributed random variables with the mean value less than one for all \(i\). It occurs in many fields (financial mathematics, hydrology, insurance and Hoare's selection algorithm). The paper presents an algorihm generating (pseudo)random numbers with the distribution of \(Y\). The case when the \(W_i\) have the distribution of \(U**(1/b)\), \(b>0\), where \(U\) is uniformly distributed on the interval \(\langle 0,1\rangle\), is dicussed in details and a new rejection method is presented. The algorithm is based on quite complex theoretical results. The main result states, that the proposed rejection algorithm halts with probability one and for some \(b\) the mean duration of the algorithm to generate a new random number is finite. No results on the experiments with the algorithm are mentioned.
- On the properties of a Takács distribution
- A multiplicative version of the Lindley recursion
- On strong and almost sure local limit theorems for a probabilistic model of the Dickman distribution
- Fast perfect simulation of Vervaat perpetuities
- Simulation of the present value of perpetuity in CIR interest model
- On exact sampling of stochastic perpetuities
- On exact sampling of nonnegative infinitely divisible random variables
- Complexity Questions in Non-Uniform Random Variate Generation
- A note on the approximation of perpetuities
- Appendix to ``Approximating perpetuities
- Approximating perpetuities
- Exact simulation of generalised Vervaat perpetuities
- Exact simulation of the extrema of stable processes
- Perpetuities and asymptotic change-point analysis
- Representation and simulation of multivariate Dickman distributions and Vervaat perpetuities
- On a modified version of the Lindley recursion
- Proof of the perpetuity equation
- Simulating the Dickman distribution
This page was built for publication: Simulating perpetuities
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5936989)