A model of optimal portfolio selection under liquidity risk and price impact (Q2463703): Difference between revisions

The authors propose a model of liquidity risk and price impact that adopt both theoretical point of view, where liquidity is defined in terms of the bid--ask spread and/or transaction costs associated with a trading strategy, and practical point of view, where illiquidity is often viewed as the risk that a trader may not be able to extricate himself from a position quickly when the need arises. The optimization problem is formulated as a parabolic impulse control problem with three variables (besides time) related to the cash holdings, number of stock shares and price. The main goal is to obtain a rigorous characterization result on the value function through the associated Hamilton--Jacobi--Bellman quasi--variational inequality. The impulse transaction function and the solvency constraint are nonlinear and this leads to technical difficulties. A suitable linearization of the liquidation value is constructed that yields a sharp upper bound for the value function. The viscosity characterization of the value function in terms of partial differential equation is presented.