Invariance times
arXiv:1702.01045
Abstract
On a probability space we consider two filtrations and a stopping time such that the predictable processes coincide with predictable processes on . In this setup it is well-known that, for any semimartingale , the process ( stopped "right before ") is a semimartingale.Given a positive constant , we call an invariance time if there exists a probability measure equivalent to on such that, for any local martingale , is a local martingale. We characterize invariance times in terms of the Azéma supermartingale of and we derive a mild and tractable invariance time sufficiency condition. We discuss invariance times in mathematical finance and BSDE applications.