Discounted linear exponential quadratic Gaussian control
: We describe a recursive formulation of discounted costs for a linear quadratic exponential Gaussian linear regulator problem which implies time-invariant linear decision rules in the infinite horizon case. Time invariance in the discounted case is attained by surrendering state-separability of the risk-adjusted costs. Address: Thomas J. Sargent, Hoover Institution, Stanford, California 94305; e-mail address: sargent@riffle.stanford.edu. This research was funded by grants from the National Science Foundation. Wen Fang Liu, Alex Taber, and Thomas Tallarini, Jr. provided helpful comments on a previous draft. Discounted Linear Exponential Quadratic Gaussian Control This paper formulates a version of a discounted Gaussian optimal linear regulator in which the return function is modified as suggested in [5], [6], [13]-[15] to incorporate a riskadjustment. In [5], the problem is formulated for the undiscounted case. Contributions [2] and [15] described how recursions on a Riccati diff...