An Interpretable Neural Network for Parameter Inference
This addresses the problem of interpretability for researchers and practitioners in economics and finance, offering a method for parameter inference in nonlinear models, though it appears incremental as it builds on existing Bayesian and neural network techniques.
The paper tackles the lack of interpretability in deep neural networks for fields like economics and finance by proposing a generative neural network architecture called PENN that estimates local posterior distributions for regression model parameters, enabling visualization and inference in complex settings.
Adoption of deep neural networks in fields such as economics or finance has been constrained by the lack of interpretability of model outcomes. This paper proposes a generative neural network architecture - the parameter encoder neural network (PENN) - capable of estimating local posterior distributions for the parameters of a regression model. The parameters fully explain predictions in terms of the inputs and permit visualization, interpretation and inference in the presence of complex heterogeneous effects and feature dependencies. The use of Bayesian inference techniques offers an intuitive mechanism to regularize local parameter estimates towards a stable solution, and to reduce noise-fitting in settings of limited data availability. The proposed neural network is particularly well-suited to applications in economics and finance, where parameter inference plays an important role. An application to an asset pricing problem demonstrates how the PENN can be used to explore nonlinear risk dynamics in financial markets, and to compare empirical nonlinear effects to behavior posited by financial theory.