Generalized Distribution Prediction for Asset Returns
This provides a more accurate and generalizable method for financial analysts and investors to forecast return distributions across various asset classes, though it is incremental as it builds on existing quantile and LSTM techniques.
The paper tackles predicting asset return distributions by developing a two-stage quantile-based LSTM model that incorporates asset-specific and market data, achieving a 98% improvement over linear quantile regression and over 50% better than dense neural networks.
We present a novel approach for predicting the distribution of asset returns using a quantile-based method with Long Short-Term Memory (LSTM) networks. Our model is designed in two stages: the first focuses on predicting the quantiles of normalized asset returns using asset-specific features, while the second stage incorporates market data to adjust these predictions for broader economic conditions. This results in a generalized model that can be applied across various asset classes, including commodities, cryptocurrencies, as well as synthetic datasets. The predicted quantiles are then converted into full probability distributions through kernel density estimation, allowing for more precise return distribution predictions and inferencing. The LSTM model significantly outperforms a linear quantile regression baseline by 98% and a dense neural network model by over 50%, showcasing its ability to capture complex patterns in financial return distributions across both synthetic and real-world data. By using exclusively asset-class-neutral features, our model achieves robust, generalizable results.