Researchers have developed a new model called the Consensus-Bottleneck Asset Pricing Model (CB-APM) designed for predicting stock returns. This model integrates aggregate analyst consensus as a key component, treating professional beliefs as a proxy for market information. The CB-APM aims for interpretability by design, using its bottleneck structure to improve predictive accuracy and focus on economically meaningful drivers. Portfolios based on CB-APM forecasts have shown strong, consistent returns across different economic conditions, and the model identifies priced risk variations missed by traditional factor models. AI
Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →
IMPACT Introduces a novel interpretable deep learning approach for asset pricing, potentially improving quantitative trading strategies.
RANK_REASON Academic paper introducing a new model for financial forecasting.