STAN: Smooth Transition Autoregressive Networks
This work addresses forecasting challenges in economics and finance, but it is incremental as it adapts an existing model to a neural network framework.
The paper tackled the problem of modeling regime-dependent dynamics in economic and financial forecasting by proposing a neural network architecture that mimics Smooth Transition Autoregressive (STAR) models, resulting in a more flexible and scalable approach with potential for enhanced predictive accuracy.
Traditional Smooth Transition Autoregressive (STAR) models offer an effective way to model these dynamics through smooth regime changes based on specific transition variables. In this paper, we propose a novel approach by drawing an analogy between STAR models and a multilayer neural network architecture. Our proposed neural network architecture mimics the STAR framework, employing multiple layers to simulate the smooth transition between regimes and capturing complex, nonlinear relationships. The network's hidden layers and activation functions are structured to replicate the gradual switching behavior typical of STAR models, allowing for a more flexible and scalable approach to regime-dependent modeling. This research suggests that neural networks can provide a powerful alternative to STAR models, with the potential to enhance predictive accuracy in economic and financial forecasting.