A general stochastic calculus approach to insider trading (Q2493284)

The authors consider a market driven by a standard Brownian motion \(B_t\) on a filtered probability space, where the coefficients are adapted to a filtration \(\mathbb{G}\). The insider in this market is a person who has access to a filtration \(\mathbb{H}\), which is strictly bigger than \(\mathbb{G}\), but it is not supposed that \(B_t\) is a semimartingale with respect to \(\mathbb{H}\). Instead of this, an insider strategy is represented by an \(\mathbb{H}_t\)-adapted process \(\varphi(t)\) and all anticipating integrals are interpreted as the forward integrals. An optimal portfolio problem with general utility for an insider is considered and it is shown that if an optimal insider portfolio \(\pi^*(t)\) of this problem exists, then \(B_t\) is an \(\mathbb{H}_t\)-semimartingale, i.e., the enlargement of filtration property holds. This is a converse to previously known results in this field. Moreover, if \(\pi^*\) exists, an explicit expression in terms of \(\pi^*\) for the semimartingale decomposition of \(B_t\) with respect to \(\mathbb{H}_t\) is obtained.