In this paper we will investigate a model of how banks allocate assets to manage market, credit, and liquidity risk. Many models consider the role of reserve supply in the bank’s optimization prob- lem, here we include a shock in reserve supply to implement liquidity risk. The solution is com- pleted numerically, utilizing value function iteration with Quasi-Monte Carlo to handle stochastic quantities. The solution method is able to model liquidity risk with a wide range of feasible shocks to give a more realistic outcome than previous models. The algorithm created is general enough to allow for easy parallelization to increase computation speed, as well as applications to other high dimensional stochastic dynamic programming problems.
