The Method of Moderation
Consumer spending lies between pessimistic and optimistic extremes. This paper uses that fact to redefine the search space for optimal consumption rules.
Consumption
Theory-Informed Neural Networks for Heterogeneous-Agent Macroeconomics: A Generalized Method of Moderation for Consumption under Uncertainty
Analytical buffer-stock bounds in the network’s output layer anchor the learned policy in the high-wealth tail. On the Krusell-Smith testbed, out-of-sample Euler errors are orders of magnitude smaller where capital concentrates.