Abstract: We discuss problem formulations and equilibrium solutions to time-inconsistent consumption-investment problems. In one problem class, we formalize preference uncertainty and discuss solutions dependent on learning and preferences regarding preference risk.

In another problem class, we suggest time-global disentanglements of time and risk preferences as alternatives to the classical time-local recursive utility. We compare the results, formally and numerically, to those of the recursive utility and discuss extensions. The newest suggestion is the time-continuous Kihlstrom-Mirman preferences.