Practical considerations for regression methods for stochastic control problems with utility functions

The focus of this talk is the application of regressions methods to stochastic control and stopping problems involving utility functions and wealth dynamics, a common use-case in the literature: we discuss our methodology regarding the choice of basis functions, regression model, and optimisation approach. We illustrate our approach with a stochastic control and stopping model for optimal residential energy consumption and timing of an energy-efficiency investment.