A nonrandom variational approach to stochastic linear quadratic Gaussian optimization involving fractional noises (FLQG) (Q2574323)

The continuous-time LQG problem is revisited. Main novelty is a fractional Brownian motion which is used to model an additive perturbation in the state equation. The problem is solved completely using a non-random dynamical equations of the moments and fractional Taylor's series. The Taylor's expansion of fractional order is used to explain relation between derivatives and fractional derivatives. Maruyama notation of the fractional order is used and the general framework for the approximate optimization is described. Moreover the LQG problem with the fractional state dependent noise is considered and Riccati type equation is obtained, solution of which is used to find the optimal control law. As an example the author solves the optimal portfolio problem and the optimal policy obtained seems to be similar to the one used by investors in practice.