Two chain-transformations and their applications to quantiles
If \(M_\alpha(X)\) denotes the quantile associated with the occupation measure of a real valued stochastic process \(X= (X_s, 0\leq s\leq 1)\), the following identity (in distribution) has been proved in previous works (it is called Wendel-Port-Dassios identity by the authors): \[ M_\alpha(X)= \sup_{0\leq t\leq \alpha} X_t+ \inf_{0\leq t\leq 1-\alpha} (X_{t+ \alpha}- X_\alpha).\tag{1} \] The aim of the paper is to obtain this identity by path transformations. The authors describe two chain transformations, one of predictable kind, the other of an optional kind; these transformations are used to prove (1) for renewal reward processes and for chains with exchangeable increments, respectively. Continuous time results for Lévy processes are also obtained, using Meyer-Tanaka formula for local time. In the case of Lévy processes with no negative jumps, explicit formulae are also proved.
- A proof of Dassios' representation of the \(\alpha\)-quantile of Brownian motion with drift
- Sample quantiles of stochastic processes with stationary and independent ents
- Pricing and hedging of quantile options in a flexible jump diffusion model
- Sample quantiles of additive renewal reward processes
- A Path Transformation and its Applications to Fluctuation Theory
- An extension of Vervaat's transformation and its consequences
- The quantile transform of simple walks and Brownian motion
This page was built for publication: Two chain-transformations and their applications to quantiles
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4376510)
