A News-based Machine Learning Model for Adaptive Asset Pricing
This addresses asset pricing for investors by providing an incremental improvement over existing factor models.
The paper tackled the problem of predicting stock returns by introducing the News Embedding UMAP Selection (NEUS) model, which uses financial news and machine learning to select basis assets, resulting in significantly better fitting and prediction power than the Fama-French 5-factor model.
The paper proposes a new asset pricing model -- the News Embedding UMAP Selection (NEUS) model, to explain and predict the stock returns based on the financial news. Using a combination of various machine learning algorithms, we first derive a company embedding vector for each basis asset from the financial news. Then we obtain a collection of the basis assets based on their company embedding. After that for each stock, we select the basis assets to explain and predict the stock return with high-dimensional statistical methods. The new model is shown to have a significantly better fitting and prediction power than the Fama-French 5-factor model.