An approximation procedure for asymmetric stable Paretian densities (Q1297871)

The authors propose a two-step polynomial-based procedure for approximating the probability density function of the asymmetric stable Paretian distribution. This approach is straightforward and requires solution of systems of linear equations. In an illustrative application of the procedure it reduces the computational burden by a factor of about ten when compared to the fast Fourier transform based approximation. The proposed approximation is easily implemented and leads to fast evaluation of the probability density function as typically required in maximum-likelihood estimation or in the online decision making.