Optimal Exchange Rate Policy Under Unknown Pass-through and Learning With Applications to Korea

Abstract

The purpose of this analysis is to explore optimal monetary and exchange rate policy in continuous time when there is uncertainty in the exchange rate pass-through mechanism. Selecting official reserves as an operating target, optimal feedback control rules are derived using MATLAB Simulink under a stochastic linear-quadratic Gaussian specification with a Kalman observer approach to learning the differing pass-through expectations. The model is estimated using discrete time Korean data, which is transformed into the continuous time model where simulations are applied to the Korean economy.