Feature Engineering Methods on Multivariate Time-Series Data for Financial Data Science Competitions
Feature engineering methods applied to US market price data for financial predictions in Numerai-Signals competition.
What it examines
This paper explores various feature engineering methods applied to US market price data for financial data science competitions, focusing on creating robust machine learning models to predict stock rankings in the Numerai-Signals tournament.
What it concludes
The research underscores the effectiveness of feature engineering in financial time series prediction, particularly using sentiment data. Future work could explore more diverse data sources and advanced feature extraction methods to enhance predictive accuracy.
Evidence objects
The research underscores the effectiveness of feature engineering in financial time series prediction, particularly using sentiment data. Future work could explore more diverse data sources and advanced feature extraction methods to enhance predictive accuracy.
key_findings bullet 1 · key_findings · validation V0
Raw abstract and provenance
Abstract: This paper is a work in progress. We are looking for collaborators to provide us financial datasets in Equity/Futures market to conduct more bench-marking studies. The authors have papers employing similar methods applied on the Numerai dataset, which is freely available but obfuscated. We apply different feature engineering methods for time-series to US market price data. The predictive power o… ▽ More This paper is a work in progress. We are looking for collaborators to provide us financial datasets in Equity/Futures market to conduct more bench-marking studies. The authors have papers employing similar methods applied on the Numerai dataset, which is freely available but obfuscated. We apply different feature engineering methods for time-series to US market price data. The predictive power of models are tested against Numerai-Signals targets. △ Less
Source row: 792 · abstract type: unknown