HMM-LSTM Fusion Model for Economic Forecasting

TL;DR

This work tackles CPI inflation forecasting under regime shifts by fusing Hidden Markov Model (HMM) regime features with Long Short-Term Memory (LSTM) networks. It augments LSTM inputs with HMM-derived hidden states and state means and uses Integrated Gradients to provide interpretability for a typically black-box model. The results show improved short-term predictive accuracy and coherent attribution of drivers, though longer horizons exhibit more limited gains. The approach bridges traditional economic regime analysis with deep learning to produce more robust, interpretable forecasts with potential policy relevance.

Abstract

This paper explores the application of Hidden Markov Models (HMM) and Long Short-Term Memory (LSTM) neural networks for economic forecasting, focusing on predicting CPI inflation rates. The study explores a new approach that integrates HMM-derived hidden states and means as additional features for LSTM modeling, aiming to enhance the interpretability and predictive performance of the models. The research begins with data collection and preprocessing, followed by the implementation of the HMM to identify hidden states representing distinct economic conditions. Subsequently, LSTM models are trained using the original and augmented data sets, allowing for comparative analysis and evaluation. The results demonstrate that incorporating HMM-derived data improves the predictive accuracy of LSTM models, particularly in capturing complex temporal patterns and mitigating the impact of volatile economic conditions. Additionally, the paper discusses the implementation of Integrated Gradients for model interpretability and provides insights into the economic dynamics reflected in the forecasting outcomes.

Figures (18)

• Figure 1: Economic Features of the US
• Figure 2: Correlation Matrix of the Variables
• Figure 3: Transmission Probabilities of HMM
• Figure 4: HMM to Recessions
• Figure 5: HMM to Recessions Transformed