Kavout: K Score-Dynamic Alpha Signal By Machine Learning – US Equities
Machine learning US Equities
Prediction power from weeks to 2-3 months in the future, NOT a quantitative relation to explain the past performance.
Evaluated over 1000 factors, and defined over 200 factors in house as input to train proprietary machine learning model.
Simplified rating that incorporated dynamic factor shifting and rotation.
Predictive equity rating from 1-9 for few weeks to 2-3 months in the future with incorporation of dynamic factor shifting. The development of K Scores takes in fundamental information, pricing and trading volumes, technical signals, and alternative data such as sentiment, and apply the best-of-breed proprietary learning algorithms, and uses methods such as regression, classification, model selection, ranking algorithms, deep learning and more to rank stocks. Our intelligent system identifies the intrinsic relationships, uncovers the dependency structure and calculates predictive analytics for future performance.